David S. Tait

Lesions of the dorsal noradrenergic bundle impair attentional set‐shifting in the rat

Rats with medial prefrontal cortex (mPFC) lesions are impaired in attentional set‐shifting, when it is required to shift to a previously irrelevant perceptual dimension. The main source of noradrenergic input to the mPFC is from the locus coeruleus via the dorsal noradrenergic ascending bundle (DNAB). This study examined the effects of selective cortical noradrenaline depletion following 6‐hydroxydopamine‐induced lesions of the DNAB on attentional set‐shifting and other aspects of discrimination learning and performance. Rats learned to dig in baited bowls, and then acquired discriminations based on one of two aspects of a bowl − odour or digging medium. The task tested acquisition of novel discriminations (both intra‐ and extra‐dimensional) and reversal learning when contingencies were reversed with the same stimuli. At the conclusion of testing, the DNAB‐lesioned rats were shown to have a selective depletion of noradrenaline of ∼ 70% within the mPFC (cingulate and prelimbic cortex subregions), with no other significant changes in dopamine or 5‐hydroxytryptamine. Rats required more trials to learn new discriminations when attentional shifting was required [extra‐dimensional (ED)‐shift]. Rats with dorsal noradrenergic ascending bundle (DNAB) lesions were impaired in novel acquisitions when an ED‐shift was required, but were unimpaired in reversal learning and other aspects of discrimination learning, relative to controls. These data are consistent with other evidence implicating noradrenaline (NA) in attentional set‐shifting, and contrast with effects of manipulations of 5‐hydroxytryptamine (5‐HT) and acetylcholine within the medial prefrontal cortex (mPFC). The findings are also relevant to recent theorizing about the functions of the coeruleo‐cortical noradrenergic system.

Difficulty Overcoming Learned Non-reward during Reversal Learning in Rats with Ibotenic Acid Lesions of Orbital Prefrontal Cortex

Abstract: Behavioral flexibility is a concept often invoked when describing the function of the prefrontal cortex. However, the psychological substrate of behavioral flexibility is complex. Its key components are allocation of attention, goal‐directedness, planning, working memory, and response selection. Furthermore, there is evidence that different regions of the prefrontal cortex might be implicated in these different components. In rule‐switching tasks, a distinction is made between errors that are perseverative (difficulty switching from a previously rewarded strategy) and errors due to learned‐irrelevance (difficulty switching to a strategy previously uncorrelated with reward). A similar distinction might be made for reversal learning, which involves inhibition of a previously rewarded response and activation of a previously unrewarded response. Damage to the orbital prefrontal cortex (OPFC) results in a deficit in reversal learning. The present study was designed to examine whether one or both of either perseveration or learned non‐reward might account for the deficit. Rats with bilateral ibotenic acid‐induced lesions of the OPFC were not impaired in acquisition of discriminations. They were impaired, relative to controls, only when they had to overcome learned non‐reward. They did not show enhanced perseveration. We conclude that an inability to overcome learned non‐reward significantly contributes to OPFC lesion‐induced deficits in behavioral flexibility.

Lesions of the basal forebrain impair reversal learning but not shifting of attentional set in rats

The cholinergic neurons of the basal forebrain, which project to cortex, the thalamic reticular nucleus and the amygdala, are implicated in many aspects of attentional function, while the intrinsic neurons of the basal forebrain are implicated in learning and memory. This study compared the effects of lesions of the basal forebrain made with either the immunotoxin 192-IgG-saporin (which selectively destroys cholinergic neurons), or the non-selective excitotoxin, ibotenic acid (which destroys both cholinergic and non-cholinergic neurons) on a task which measure the acquisition and shifting of attentional set as well as the ability to learn reversals of specific stimulus-reward pairings. Rats learned to obtain food reward by digging in small bowls containing distinctive digging media that were differentially scented with distinct odours. They performed a series of two-choice discriminations, with the bait associated with either the odour or the digging medium. Rats with 192-IgG-saporin lesions of the basal forebrain were not impaired relative to control rats at any stage of the task. Rats with ibotenic acid lesions of the basal forebrain were impaired the first time stimulus-reward contingencies were reversed. They were not impaired in acquisition of new discriminations, even when an attentional-shift was required. These data are consistent with data from marmosets and so highlight the functional similarity of monkey and rodent basal forebrain. They also confirm the likely involvement of non-cholinergic neurons of the basal forebrain in reversal learning.

Lesions of the orbital prefrontal cortex impair the formation of attentional set in rats.

In rats, reversal learning impairments are commonly reported after lesions of the orbital prefrontal cortex (OFC), in contrast to the effect of lesions of the medial prefrontal cortex, which impair attentional set‐shifting. Comparable dissociations have also been reported in humans, monkeys and mice. However, these two manifestations of behavioural flexibility may share common cognitive processes. The present study tested the hypothesis that lesions of the OFC (an area that integrates expected and actual outcomes to signal which cues in the environment predict reward) would impair the formation of attentional set as well as impairing reversal learning. We compared the performance of lesioned and control rats on two set‐shifting tasks. The first task we used, ‘the 4ID task’, had no reversal stages, but multiple intradimensional acquisitions before the extradimensional shift stage, to assess set‐formation as well as set‐shifting. The second task was the standard intradimensional/extradimensional ‘7‐stage task’, which includes reversal learning stages after each compound acquisition. Compared with controls, lesioned rats were slower to form attentional set on the 4ID task. When they did form a set, they required more trials to complete the extradimensional shift stage. On the 7‐stage task, we replicated our previous finding of impaired reversal learning and reduced shift‐costs. We interpret these findings as reflecting a single deficit in identifying relevant cues after unexpected outcomes, which supports recent models of OFC function. Our findings challenge the assumption that the contribution of the OFC to behavioural flexibility is limited to reversal learning.

Lesions of the dorsomedial striatum impair formation of attentional set in rats.

Behavioural flexibility refers to the ability to rapidly adapt to novel situations and it has been suggested that the frontal lobe and basal ganglia are implicated in various components of adjusting to changes in environmental contingencies. Behavioural flexibility can be assessed using attentional set-shifting tasks, in which performance is impaired after damage to the prefrontal cortex. The present study explores the downstream contribution of the prefrontal projection zone in the dorsomedial striatum (DMS) to attentional set shifting. Rats were tested in two set-shifting tasks following quinolinic acid injections bilaterally into the DMS. When tested in a rodent version of the set-shifting task, rats with a DMS lesion displayed a greater number of errors during the reversal stages of the task than sham lesion controls but the nature of the errors did not differ between the two groups. Interestingly, when the rats were tested in a modified version of the set-shifting task, directly designed for measuring the formation of an attentional set, sham lesion controls displayed a pronounced shift-cost, evident of successful set-formation. In contrast, rats with DMS lesions failed to form an attentional set, showing no performance cost when a shift of attention was required. These results support previous reports of the importance of the DMS in behavioural flexibility but also suggest that this region is vital for the formation of set, possibly by extrapolating different perceptions into a unified representation of a dimension.

Attentional Set-Shifting in Rodents: A Review of Behavioural Methods and Pharmacological Results

Attentional set-shifting tasks have been used as a measure of human fronto-executive function for over 60 years. The major contribution these tasks have made has been the quantification of cognitive deficits associated with human pathologies such as schizophrenia, attention deficit/hyperactivity disorder and dementias related to Parkinsons, Huntingtons and Alzheimers diseases. Thirteen years ago an intradimensional/extradimensional attentional set-shifting task was developed for rats. Since then, there have been over 70 publications detailing the effects of various manipulations on task performance in rats, and 17 publications describing adaptations of the task for mice. Much of this literature has focused on animal models of neuropathology and cognitive deficits associated with schizophrenia and other human conditions. Altogether, these results have elucidated the roles of multiple neurotransmitters in the manifestation of cognitive deficits, and their subsequent amelioration, including dopamine, serotonin, acetylcholine and noradrenaline. However, the fundamental promise of the attentional set-shifting task, to measure cognitive flexibility in humans and rodents in a formally analogous way, has often been under investigated and over simplified. This review explores the research that led to the development of the rat attentional set-shifting task, and how subsequent use of the task has expanded our understanding of the psychological and neurological underpinnings of discrimination and reversal learning, as well as the formation, maintenance and shifting of attentional set.

Attentional Set-Shifting Across Species.

Attentional set-shifting, as a measure of executive flexibility, has been a staple of investigations into human cognition for over six decades. Mediated by the frontal cortex in mammals, the cognitive processes involved in forming, maintaining and shifting an attentional set are vulnerable to dysfunction arising from a number of human neurodegenerative diseases (such as Alzheimers, Parkinsons and Huntingtons diseases) and other neurological disorders (such as schizophrenia, depression, and attention deficit/hyperactivity disorder). Our understanding of these diseases and disorders, and the cognitive impairments induced by them, continues to advance, in tandem with an increasing number of tools at our disposal. In this chapter, we review and compare commonly used attentional set-shifting tasks (the Wisconsin Card Sorting Task and Intradimensional/Extradimensional tasks) and their applicability across species. In addition to humans, attentional set-shifting has been observed in a number of other animals, with a substantial body of literature describing performance in monkeys and rodents. We consider the task designs used to investigate attentional set-shifting in these species and the methods used to model human diseases and disorders, and ultimately the comparisons and differences between species-specific tasks, and between performance across species.

Tacrine improves reversal learning in older rats

Age-related decline has been reported in most cognitive domains, including executive function: in particular, attentional set-shifting and reversal learning, as measures of executive control, are impaired in aged populations of both humans and rats. Despite the importance of the cholinergic system in age-related cognitive decline, no data are available on the effects of cholinergic enhancement on age-related performance deficits in tests of attentional set-shifting. We investigated the effects of the cholinesterase inhibitor tetrahydroacridin-9-amine (tacrine) on reversal learning and attentional set-shifting in older rats (aged 16-21 months) using the rodent version of the intradimensional/extradimensional attentional set-shifting task in a repeated-measures design. Discrimination acquisition was not impaired, but age-related impairments in reversal learning were persistent between tests, and ameliorated by the 3 mg/kg dose of tacrine. No age-related impairments in set-shifting were seen, but there was a tendency for tacrine to reduce the cost of shifting set. Given the lack of previous evidence for a role of cortical acetylcholine in attentional set-shifting tasks, it is likely that altered neurotransmitter interactions in striatum underlie this improvement.

The Concerted Activity in Parallel Proprioceptive Pathways Controls the Initiation of Co-activation in the Locust Kick Motor Programme

To jump and kick the locust uses a catapult mechanism implemented by a three‐stage motor programme: initial flexion of the hind tibiae, co‐activation of the antagonist flexor and extensor tibiae motor neurons and trigger inhibition of the flexor motorneurons. The transition from stage 1 to stage 2 thus involves a switch from the normal alternate activation to co‐activation of the antagonist tibiae motorneurons. However, co‐activation has never been observed when the central nervous system has been isolated from the leg. This led us to investigate the possibility that the transition from stage 1 to stage 2 is controlled by a proprioceptive signal. In the first set of experiments intracellular recordings were made in the flexor and extensor motorneurons while the position of the tendon of the femoral chordotonal organ (FCO), which signals tibial position and movement, was experimentally controlled. In these heavily dissected preparations, stretch of the FCO tendon (signalling tibial flexion) was a necessary condition for co‐activation. However, in minimally dissected preparations (in which merely EMG recordings were made), we found that co‐activation occurred even when the FCO was signalling tibial extension, suggesting the involvement of other proprioceptors. A series of experiments were then conducted on minimally dissected preparations to determine the relative contributions of each of the three main hind leg proprioceptors which might signal tibial flexion: the FCO, the lump receptor and Brünners organ. When all three proprioceptors were intact the chance of evoking co‐activation was largest, when all three were eliminated co‐activation could no longer be evoked, irrespective of the level of arousal. Various combinations of partial de‐afferentation showed that the FCO plays the major role, with the lump receptor and Brünners organ playing significant, but progressively less important, roles. We conclude that the three receptors act together as a permissive proprioceptive gate for the kick and jump motor programme, but with a hierarchy of the strengths of their effectiveness.

Effects of lesions of the subthalamic nucleus/zona incerta area and dorsomedial striatum on attentional set-shifting in the rat.

Highlights • Reversal learning impaired after dorsomedial striatal dopamine depletion.• No evidence of set-formation after excitotoxic subthalamic nucleus lesions.• No impairments observed after both lesions together.