Spatial sequence memory and spatial error monitoring in the Groton Maze Learning Task (GMLT): A validation study of GMLT sub-measures in healthy children

The stepping-stone variant of the hidden pathway maze learning (HPML) task paradigm has been extensively used to investigate cognitive functions in neuropsychology and neuropharmacology. Previous studies have used total error across trials, as well as rule-break errors and learning errors, to define spatial memory and/or executive function in healthy and impaired adults and children. However, the construct validity of performance measures on HPML tasks has not been established in healthy children. To assess the construct validity of measures of exploratory and rule-break errors on the Groton Maze Learning Task (GMLT) measures of spatial sequence memory (Corsi Blocks Task) and spatial error monitoring (Continuous Paired Associate Learning; CPAL) were used. The results indicate that Corsi span predicted GMLT spatial sequence memory and CPAL accuracy predicted GMLT spatial error monitoring. The construct validity of the GMLT as a measure of spatial memory and executive function are discussed with regard to prior research using HPML tasks in neuropsychological contexts.     

Water can induce better spatial memory performance than food in radial maze learning by rats1

Abstract: Two experiments were carried out to test the differential effects of hunger and thirst on memory performance. In Experiment 1, two groups of rats were exposed to an original radial‐maze task and then to a 30‐min retention‐memory task. The food‐deprived group completed the original task more quickly than the water‐deprived group, but the thirsty group mastered the memory task more quickly than the hungry group (p < 0.01). In Experiment 2, deprivation conditions were changed from the original to the memory task. The food‐water group completed the memory task more rapidly than the water‐food group (p < 0.05). Thirst proved to constitute a more favorable condition for retention‐memory learning. The applicability of several theories is discussed.     

Hippocampus-dependent strategy choice predicts low levels of cell proliferation in the dentate gyrus

Neurogenesis continues to occur throughout life in the mammalian hippocampus. Previous research has suggested that the production of new neurons in the hippocampus during adulthood may be related to hippocampus-dependent learning and memory. However, the exact relationship between adult neurogenesis and learning and memory remains unclear. Here we investigated whether learning strategy selection is related to cell proliferation or to survival of new neurons in the hippocampus of adult male rats. We trained rats on alternating blocks of hippocampus-dependent (hidden platform) and hippocampus-independent (visible platform) versions of the Morris water task with the platform always in the same position. Following training, rats were given a probe session during which the platform was visible and in a novel location. Preferred strategy was determined by observing the initial swim path. Rats were classified as place strategy (hippocampus-dependent) users if they swam to the old platform location. Cue strategy (hippocampus-independent) users were classified as those rats that swam initially to the visible platform. Our results indicate that rats that preferentially used a place strategy had significantly lower cell proliferation than cue strategy users. However, there was no significant difference in cell survival or number of immature neurons between strategy user groups. These results suggest that low levels of cell proliferation in the dentate gyrus may be conducive or coincident with more efficient memory processing in the hippocampus.     

Damage to the retrosplenial cortex produces specific impairments in spatial working memory

Mounting evidence indicates that the retrosplenial cortex (RSP) has a critical role in spatial navigation. The goal of the present study was to characterize the specific nature of spatial memory deficits that are observed following damage to RSP. Rats with RSP lesions or sham lesions were first trained in a working memory task using an 8-arm radial arm maze. Rats were allowed 5 min to visit each arm and retrieve food pellets and a 5-s delay was imposed between arm choices. Consistent with previous research, rats with RSP damage committed more errors than controls. In particular, RSP-lesioned rats committed more errors of omission (failing to visit an arm of the maze), but there were no lesion effects on errors of commission (revisiting an arm). Neither group of rats exhibited a turn bias (i.e., always turning a certain direction when choosing an arm). At the end of the training phase of the experiment, both groups had reached asymptote and committed very few errors. In the subsequent test phase, a longer delay (30-s) was imposed during some sessions. Both control and RSP-lesioned rats continued to make few errors during sessions with the standard 5-s delay, but RSP-lesioned rats were impaired at the 30-s delay and committed more errors of commission, consistent with an increase in taxing spatial working memory.     

Effect of exposure to lithium-paired or amphetamine-paired saccharin solution on open arm avoidance in an elevated plus maze

Recent evidence suggests that drug-induced conditioned taste avoidance may be mediated by conditioned fear (e.g., Parker, 2003). The experiments reported here evaluated the effect of exposure to a drug-paired flavor on open arm exploration in an elevated plus maze (EPM), a measure of fear. When rats were tested on a familiar (trial 2) EPM, but not on a novel (trial 1) EPM, prior exposure to a lithium-paired saccharin solution enhanced open arm activity relative to saline-paired saccharin. On the other hand, when rats were exposed to lithium-paired saccharin during plus maze exposure, they displayed suppressed open arm activity relative to unpaired controls when tested in a familiar maze. The pattern of results was specific to the conditional affective properties of the taste, because exposure to unconditional sickness produced by administration of lithium, and unconditionally unpalatable quinine solution did not produce this pattern. These results were interpreted in terms of the opponent process model of motivation; that is, exposure to a lithium-paired flavor elicits conditioned fear which is immediately followed by conditioned relief when the exposure is terminated. On the other hand, exposure to an amphetamine-paired flavor either before or during EPM testing enhanced open arm exploration. Since the strength of taste avoidance did not differ among amphetamine and lithium conditioned rats, these results provide further evidence that the nature of a saccharin-lithium association differs from that of a saccharin-amphetamine association.     

Mice selectively bred for an open field activity increase after maze learning

A unidirectional selective breeding experiment performed over six generations resulted in a line of mice (S6), which differed from the maintained unselected Swiss albino strain, called normal (N) strain, in the following respects: S6 mice increased their open field activity after maze learning significantly more than N mice. S6 mice ambulated more and exhibited more thigmotactic behaviour in a circular open field than N mice. S6 mice were superior than N mice in regard to maze learning capacity. Finally, S6 mice were interpreted as significantly less emotional according to their defecation, more responsive to novelty according to their urine pattern and more aggressive than N mice.     

Radial maze analog for pigeons: Evidence for flexible coding strategies may result from faulty assumptions

Previous research with the radial maze has found evidence that rats can remember both places that they have already been (retrospective coding) and places they have yet to visit (prospective coding; Cook, R. G., Brown, M. F., & Riley, D. A. (1985). Flexible memory processing by rats: Use of prospective and retrospective information in the radial maze. Journal of Experimental Psychology: Animal Behaviour Processes, 11, 453-469). Such dual coding also has been found in pigeons using a radial maze analog in which insertion of a delay at different points during a trial affects performance differentially depending on where in the trial it is inserted. When a delay is interpolated either early or late in a trial, there is minimal disruption of performance compared with when it is interpolated in the middle of the trial. However, the analysis required with this procedure requires the assumption that if errors made on control trials are subtracted from errors made in delay trials, the remaining errors can be directly attributed to the delay. But errors may also be attributed to the changing criterion for making a response as the trial proceeds. Furthermore, the animal’s tendency to choose alternatives in a systematic order may also affect its need to remember the sequence of choices made (and yet to be made) on each trial. In the present research, we avoided having to make this assumption by giving the pigeons a two-alternative choice at the time of testing and by randomly determining for the pigeon the order of predelay choices on each trial. This change in procedure resulted in comparable performance as a function of where in the trial the test occurred on both control and delay trials. The effect of the delay was to produce a general decrement in performance independent of where it occurred in the trial.     

Emergence of an egocentric cue guiding and allocentric inferring strategy that mirrors hippocampal brain-derived neurotrophic factor (BDNF) expression in the Morris water maze

From insects to humans, successful navigation relies on retained representations of spatial relations. These representations are thought to depend on the hippocampal formation, particularly those that are independent of the navigator (allocentric representations). The Morris water maze is a simple and popular task often used to assess spatial navigation. But how animals navigate toward and retain information regarding the location of the goal in this task remains unclear. We provide a comprehensive account of how the water maze is accomplished behaviourally. Our findings suggest that animals solve the task using distal cues via an initial view-matching strategy that is supported by egocentric guidance. Through increased training, however, an emergence of an egocentric-guiding strategy combined with the animal’s greater ability to infer the hidden platform’s location (via allocentric extrapolation) emerges. We also demonstrate that behavioural changes, towards a more allocentric strategy, are reflected in increases in hippocampal brain-derived neurotrophic factor.     

The ameliorating effect of oroxylin A on scopolamine-induced memory impairment in mice

Oroxylin A is a flavonoid and was originally isolated from the root of Scutellaria baicalensis Georgi., one of the most important medicinal herbs in traditional Chinese medicine. The aim of this study was to investigate the ameliorating effects of oroxylin A on memory impairment using the passive avoidance test, the Y-maze test, and the Morris water maze test in mice. Drug-induced amnesia was induced by administering scopolamine (1 mg/kg, i.p.) or diazepam (1 mg/kg, i.p.). Oroxylin A (5 mg/kg) significantly reversed cognitive impairments in mice by passive avoidance and the Y-maze testing (P<.05). Oroxylin A also improved escape latencies in training trials and increased swimming times and distances within the target zone of the Morris water maze (P<.05). Moreover, the ameliorating effects of oroxylin A were antagonized by both muscimol and diazepam (0.25 mg/kg, i.p., respectively), which are GABA(A) receptor agonists. Furthermore, oroxylin A (100 microM) was found to inhibit GABA-induced inward Cl(-) current in a single cortical neuron. These results suggest that oroxylin A may be useful for the treatment of cognitive impairments induced by cholinergic dysfunction via the GABAergic nervous system.     

Contributions of the dorsal hippocampus and the dorsal subiculum to processing of idiothetic information and spatial memory

It is well established that the dorsal hippocampal formation is crucial for spatial memory in rats. However, little is known about the distinct functions of the dorsal hippocampus and the dorsal subiculum. To examine the role of the dorsal hippocampus and the dorsal subiculum, Long-Evans rats with excitotoxic lesions (NMDA) of the dorsal hippocampus (DH), the dorsal subiculum (DS), or both (DHDS), and controls were trained on a nonmatching-to-place task. Then, to identify the strategy used by each group, they were tested on the same task in the dark with the T-maze being rotated between the sample and the test runs. In the light, rats with combined lesions were impaired. In the dark, groups DH, DS, and controls performed near chance level except in trials without rotation, suggesting the use of a sense of direction. The same rats were trained on a radial-arm maze task. In the light, where proximal visual cues were accessible, combined lesions affected performance whereas in the dark, it was impaired by all lesions. This experiment demonstrated that the dorsal subiculum and the dorsal hippocampus play a crucial role in processing idiothetic information and/or in maintenance of this information in memory.