On the behavioral significance of head direction cells: neural and behavioral dynamics during spatial memory tasks.

Current theories assume that rats use the directional information reflected by head direction (HD) cells when performing spatial tasks. This assumption was assessed by monitoring anterior thalamic HD cell activity and relating it to the subject's behavioral response on 2 spatial memory tasks that tested either reference memory or working memory. In both tasks, there was a significant number of trials where there was not a tight coupling between the preferred firing direction of HD cells and the direction of the behavioral response. In addition, it was possible to intentionally change the preferred direction of HD cells without affecting performance accuracy. An additional experiment showed that manipulations that affected internal, but not external, cues impaired performance on the reference memory task. These findings suggest that HD cell activity was not consistently guiding the subjects' behavior on these 2 spatial tasks.

Maintenance of rat head direction cell firing during locomotion in the vertical plane.

Previous studies have identified a subset of neurons in the rat anterodorsal thalamus (ADN) that encode head direction (HD) in absolute space and may be involved in navigation. These HD cells discharge selectively when the rat points its head in a specific direction (the preferred firing direction) in the horizontal plane. HD cells are typically recorded during free movement about a single horizontal surface. The current experiment examined how HD cell firing was influenced by 1) locomotion in the vertical plane and 2) locomotion on two different horizontal surfaces separated in height. Rats were trained in a cylindrical enclosure containing a single polarizing cue card attached to the cylinder wall, covering approximately 100 degrees of arc. The enclosure contained two horizontal surfaces: the cylinder floor and an annulus around the cylinder top 76 cm above the floor. A 90 degrees vertical mesh ladder that could be affixed at any angular position on the cylinder wall allowed the rats to locomote back and forth between the two horizontal surfaces. Rats were trained to retrieve food pellets on the cylinder floor as well as climb the mesh ladder to retrieve food pellets on the annulus. HD cell activity was monitored as the rat traversed the horizontal and vertical surfaces of the apparatus. When the angular position of the mesh corresponded to the cell's preferred firing direction, the HD cells maintained their peak discharge rate as the rat climbed up the mesh, but did not fire when the rat climbed down the mesh. In contrast, when the mesh was positioned 180 degrees opposite the preferred firing direction, HD cells did not fire when the rat climbed up the mesh, but exhibited maximal firing when the rat climbed down the mesh. When the mesh was placed 90 or 270 degrees from the preferred firing direction, HD cells exhibited background firing rates during climbing up or down the mesh. While preferred firing directions were maintained between the two horizontal surfaces, peak firing rate increased significantly (approximately 30%) on the annulus as compared with the cylinder floor. These data demonstrate that HD cells continue to discharge in the vertical plane if the vertical locomotion began with the rat's orientation corresponding to the preferred firing direction. One model consistent with these data are that HD cells define the horizontal reference frame as the animal's plane of locomotion. Further, we propose that HD cell firing, as viewed within a three-dimensional coordinate system, can be characterized as the surface of a hemitorus.

Reversible inactivation of the medial septum or nucleus basalis impairs working memory in rats: a dissociation of memory and performance.

Lidocaine-induced inactivation of the medial septum immediately after training or prior to testing in a delay radial-arm maze task produced deficits in spatial working memory that reflected impaired acquisition of the task. Injection of lidocaine into the nucleus basalis magnocellularis produced a profile of behavioral changes that indicated that temporary inactivation of this structure impaired the behavioral expression of information already stored in working memory. This appears to reflect an impairment in processes that are required for performance (i.e., attention, motivation, sensorimotor function) of the task but not for retrieval of stored information. Site-specific inactivation of the basal forebrain should help to reveal the involvement of its component structures in different aspects of cognitive function.

Firing properties of rat lateral mammillary single units: head direction, head pitch, and angular head velocity.

Many neurons in the rat anterodorsal thalamus (ADN) and postsubiculum (PoS) fire selectively when the rat points its head in a specific direction in the horizontal plane, independent of the animal's location and ongoing behavior. The lateral mammillary nuclei (LMN) are interconnected with both the ADN and PoS and, therefore, are in a pivotal position to influence ADN/PoS neurophysiology. To further understand how the head direction (HD) cell signal is generated, we recorded single neurons from the LMN of freely moving rats. The majority of cells discharged as a function of one of three types of spatial correlates: (1) directional heading, (2) head pitch, or (3) angular head velocity (AHV). LMN HD cells exhibited higher peak firing rates and greater range of directional firing than that of ADN and PoS HD cells. LMN HD cells were modulated by angular head velocity, turning direction, and anticipated the rat's future HD by a greater amount of time (approximately 95 msec) than that previously reported for ADN HD cells (approximately 25 msec). Most head pitch cells discharged when the rostrocaudal axis of the rat's head was orthogonal to the horizontal plane. Head pitch cell firing was independent of the rat's location, directional heading, and its body orientation (i.e., the cell discharged whenever the rat pointed its head up, whether standing on all four limbs or rearing). AHV cells were categorized as fast or slow AHV cells depending on whether their firing rate increased or decreased in proportion to angular head velocity. These data demonstrate that LMN neurons code direction and angular motion of the head in both horizontal and vertical planes and support the hypothesis that the LMN play an important role in processing both egocentric and allocentric spatial information.

Firing properties of head direction cells in the rat anterior thalamic nucleus: dependence on vestibular input.

Vestibular information influences spatial orientation and navigation in laboratory animals and humans. Neurons within the rat anterior thalamus encode the directional heading of the animal in absolute space. These neurons, referred to as head direction (HD) cells, fire selectively when the rat points its head in a specific direction in the horizontal plane with respect to the external laboratory reference frame. HD cells are thought to represent an essential component of a neural network that processes allocentric spatial information. The functional properties of HD cells may be dependent on vestibular input. Here, anterior thalamic HD cells were recorded before and after sodium arsanilate-induced vestibular system lesion. Vestibular lesions abolished the directional firing properties of HD cells. The time course of disruption in the directional firing properties paralleled the loss of vestibular function. Arsanilate-treated rats exhibited only minor changes in locomotor behavior, which were unlikely to account for the loss of direction-specific firing. Vestibular lesions also disrupted the influence of angular head velocity on anterior thalamic single-unit firing rates. Finally, a subset of anterior thalamic neurons recorded from vestibular-lesioned rats exhibited a pattern of intermittent firing bursts that were distinctly unrelated to HD. This novel anterior thalamic firing pattern has not been encountered in any vestibular-intact rat. These data suggest that: (1) the neural code for directional bearing is critically dependent on vestibular information; and (2) this loss of HD cell information may represent a neurobiological mechanism to account for the orientation and navigational deficits observed after vestibular dysfunction.

Stability of spatial working memory across the estrous cycle of Long-Evans rats.

In the adult rat, hippocampal dendritic synaptic connectivity in region CA1 fluctuates across the estrous cycle. This study examined the potential functional impact of such fluctuations on spatial working memory in regularly cycling Long-Evans rats. Rats were trained in a delayed non-match-to-sample radial-arm maze task and performance was monitored across the estrous cycle. There were no significant alterations in the acquisition or performance of the working memory task across the estrous cycle, with 1- or 4-h delays imposed between training and testing sessions. However, rats performed the task significantly more slowly on proestrus than on any other estrous cycle day under both delay conditions. These results indicate that while working memory remains stable, sensorimotor or motivational aspects of the performance of this radial-arm maze task may vary across the estrous cycle.

Vitamin E attenuates the effects of both reversible and irreversible inhibitors of high-affinity choline transport in vivo.

High-affinity choline transport (HAChT) is the rate limiting step in the synthesis of acetylcholine (ACh). The activity of HAChT and the binding of its selective inhibitor, [3H]hemicholinium-3 (HC-3) are affected by a number of exogenous and endogenous factors. Previous experiments demonstrated that Vitamin E pretreatment could prevent the decrease in HAChT and the cognitive deficits induced by the cholinotoxin AF64A [38]. To further examine this effect these experiments determined whether Vitamin E would alter the efficacy of both irreversible (AF64A) and reversible (HC-3) inhibitors of HAChT. In Experiment 1, rats were pretreated with Vitamin E (50 mg/kg), 24 h and 15 min, prior to bilateral icv injection of AF64A (0, 0.75, 1.5, or 3.0 nmol). HAChT was assessed in hippocampal synaptosomes, 14 days following surgery. Vitamin E prevented the dose-dependent AF64A-induced inhibition of HAChT in the hippocampus (HPC). In a second experiment, rats were pretreated with Vitamin E as above, and infused (icv) with the reversible inhibitor of HAChT, HC-3 (20 µg), or CSF. HAChT in the HPC was assessed 30 min, 4, 12, or 24 h after injection. HC-3 produced a significant decrease of HAChT (58%) that was maximal at 4 h and recovered by 24 h. Vitamin E significantly attenuated, but did not prevent, the inhibition of HAChT produced by HC-3. These experiments demonstrate that Vitamin E pretreatment can attenuate the effects of both reversible and irreversible inhibitors of HAChT. These data are discussed in terms of potential underlying mechanisms. It is possible that the neuroprotectant effects of Vitamin E on both reversible and irreversible inhibitors of HAChT reflect an action at the choline carrier and not an antioxidant effect.

Intraseptal flumazenil enhances, while diazepam binding inhibitor impairs, performance in a working memory task.

GABAA/benzodiazepine receptors in the medial septum modulate the activity of cholinergic neurons that innervate the hippocampus. Injection of benzodiazepine (BDZ) agonists into the medial septum impairs working memory performance and decreases high-affinity choline transport (HAChT) in the hippocampus. In contrast, intraseptal injection of the BDZ antagonist flumazenil increases HAChT and prevents the memory deficits induced by systemic BDZs. The present studies attempted to further characterize the behavioral effects of medial septal injections of flumazenil to an endogenous negative modulator of the GABAA/BDZ receptor complex, diazepam binding inhibitor (DBI). Male Sprague-Dawley rats were cannulated to study the effects of intraseptal injections of these BDZ ligands on spatial working memory, anxiety-related behaviors in the elevated plus maze, and on general locomotor activity. Intraseptal flumazenil (10 nmol/0.5 microliter) produced a delay-dependent enhancement of DNMTS performance after an 8-h, but not a 4-h, delay interval. This promnestic dose of flumazenil had no effect on locomotor activity and did not produce changes in measures of anxiety on the plus maze. Intraseptal injection of DBI had no effect (8 nmol/0.5 microliter) or slightly impaired (4 nmol/0.5 microliter) DNMTS radial maze performance following an 8-h delay, without producing changes in locomotion or plus maze behavior. These data demonstrate that flumazenil has a unique profile of activity in enhancing working memory following intraseptal injection.

Spatial working memory is preserved in rats treated with anabolic-androgenic steroids.

The effects of anabolic-androgenic steroid (AAS) compounds on spatial working memory were evaluated in male rats. Thirty days of administration of a high dose of three individual AAS compounds (17 alpha-methyltestosterone, methandrostenolone, or testosterone cypionate) had no effects on spatial memory or motivation as tested on a delayed non-match-to-sample radial arm maze task. Administration of these AAS compounds at doses within the human abuse range does not impair spatial working memory in rats.

Injection of IgG 192-saporin into the medial septum produces cholinergic hypofunction and dose-dependent working memory deficits.

192-IgG saporin is an anti-neuronal immunotoxin that combines the 192 monoclonal antibody to the p75 neurotrophin receptor found on terminals and cell bodies of neurons in the cholinergic basal forebrain with the ribosome-inactivating protein saporin. Injection of 100, 237.5 or 375 ng of 192-saporin into the medial septum produced dose-related deficits in a variable-delay radial-arm maze task. 192-saporin decreased the number of correct choices and increased the number of errors in the delayed non-match to sample task. These deficits persisted throughout training and were most evident in the 375 ng group. The behavioral deficits were associated with dose-dependent decreases in pre-synaptic cholinergic parameters (ie., high affinity choline uptake) in the terminal fields of the medial septum (hippocampus, cingulate, entorhinal cortex). Choline uptake was not affected in the frontal cortex or the striatum; structures not innervated by the septum. There were no changes in regional concentrations of dopamine, serotonin, or their metabolites. Site-specific injection of IgG 192-saporin is a useful approach to explore the functions of the cholinergic basal forebrain and to model diseases of cholinergic hypofunction such as Alzheimer's disease.

Medial septal benzodiazepine receptors modulate hippocampal evoked responses and long-term potentiation.

Infusion of benzodiazepine (BDZ) receptor ligands into the medial septum (MS) produces a bidirectional modulation of spatial memory retention. The present experiments sought to determine the effects of BDZ ligands upon synaptic responses and long-term potentiation (LTP) in the dentate gyrus following electrical stimulation of the angular bundle. Intraseptal infusion of the BDZ agonist, chlordiazepoxide, decreased the amplitude of the evoked population spike and increased paired-pulse facilitation at a 150-ms interstimulus interval (ISI) in a dose-dependent manner. Intraseptal infusion of the BDZ antagonist, flumazenil (10 nmol), enhanced the amplitude of the dentate population spike and also increased paired-pulse facilitation at the 150-ms ISI. There was no effect of either BDZ receptor ligand upon the slope of the rising phase of the evoked population excitatory postsynaptic potential (pEPSP). Intraseptal flumazenil also significantly enhanced the magnitude of dentate LTP induced by high-frequency stimulation of the angular bundle. Intraseptal chlordiazepoxide failed to alter LTP induction. These results indicate that intraseptal infusion of an amnestic dose of the BDZ ligand, chlordiazepoxide, decreases the excitatory responsiveness of the dentate gyrus to its synaptic input from entorhinal cortex. In contrast, the promnestic BDZ ligand, flumazenil, enhances dentate granule cell responsivity, and facilitates synaptic plasticity in the dentate gyrus network. Taken together these data suggest that the memory impairing and memory enhancing action of these compounds may be a function of their ability to alter hippocampal physiology during a critical phase of memory. The potential role of septodentate cholinergic and GABAergic projections in the present observation is discussed.

Processing the head direction cell signal: a review and commentary.

Animals require information about their location and directional heading in order to navigate. Directional information is provided by a population of cells in the postsubiculum and the anterior thalamic nuclei that encode a very accurate, continual representation of the animal's directional heading in the horizontal plane, which is independent of the animal's location. Recent studies indicate that this signal 1) arises either in the anterior thalamic nuclei or in structures upstream from it; 2) is not dependent on an intact hippocampus; 3) receives sensory inputs from both idiothetic and landmark systems; and 4) correlates well with the animal's behavior in a spatial reference memory task. Furthermore, HD cells in the anterior thalamic nuclei appear to encode what the animal's directional heading will be about 40 ms in the future, while HD cells in the postsubiculum encode the animal's current directional heading. Both the electrophysiological and anatomical data suggest that the anterior thalamic nuclei and/or the lateral mammillary nuclei may be the sites of convergence for spatial information derived from landmarks and internally-generated cues. Current evidence also indicates that the vestibular system plays a crucial role in the generation of the HD cell signal. However, the notion that the vestibular system is the sole contributor to the signal generator is difficult to reconcile with several findings; these latter findings are better accounted for with a motor efference copy signal.

Behavioral and neurobiological alterations induced by the immunotoxin 192-IgG-saporin: cholinergic and non-cholinergic effects following i.c.v. injection.

192-IgG-Saporin is an anti-neuronal immunotoxin that combines the 192 monoclonal antibody to the p75 neurotrophin receptor found on terminals and cell bodies of neurons in the cholinergic basal forebrain with the ribosome-inactivating protein saporin. Bilateral intraventricular injection of the 192-saporin produced a variety of dose-related behavioral, neurochemical, and histological alterations in adult male rats. While both the 2 micrograms and 4 micrograms dose produced comparable cholinergic hypofunction only the high dose produced behavioral changes. Behavioral deficits induced by the 4 micrograms dose of 192-saporin induced alterations in rotorod performance and reactivity on the hot-plate which recovered over 8 weeks. In addition, the 4 micrograms dose produced a persistent impairment in the acquisition and performance of standard Morris water maze task as well as a cued version of the task. The neurobiological alterations induced by 192-saporin involved both cholinergic and non-cholinergic systems. Both doses of 192-saporin produced a 60-80% decrease in high affinity choline transport in the hippocampus and cortex without altering this parameter in the striatum. In addition, there was a significant dose-related decrease of norepinephrine in the hippocampus in the high dose group. 192-saporin did not alter the content of dopamine, serotonin, or their metabolites in any region examined. 192-saporin also produced a loss of Purkinje cells in the cerebellum. This cell type also expresses the p75 receptor and appears to be a target for intraventricular 192-saporin. This complex interplay of factors makes the i.c.v. model of 192-saporin very problematic for studying the functional properties of the cholinergic basal forebrain. However, recent data suggest that injection of 192-saporin directly into components of the cholinergic basal forebrain can be used to further elaborate the function of this brain system and to model disorders of cholinergic hypofunction such as Alzheimer's disease.

Distinct profile of working memory errors following acute or chronic disruption of the cholinergic septohippocampal pathway.

The behavioral effects of two amnestic treatments (intraseptal chlordiazepoxide (CDP) and intraventricular AF64A) were examined in a delayed-nonmatch-to-sample radial-arm maze (DNMTS) paradigm. The types of errors induced by these treatments in this working memory task were assessed to determine how acute and chronic disruptions of the medial septum affect different phases of working memory (encoding, maintenance, retrieval). Rats were initially trained to perform the DNMTS task with a 1-h delay imposed between the training and the testing sessions. The first experiment demonstrated that intraseptal injection of 30 nmoles of CDP did not produce state-dependent learning. Rats were injected immediately following training with CDP or artificial cerobrospinal fluid (CSF; drug vehicle) and then prior to testing with CDP or CSF. Injection of CDP immediately following training (CDP-CSF) impaired performance of the task regardless of whether CDP was also administered before the postdelay test (CDP-CDP). Rats infused with CDP only before the postdelay test (CSF-CDP) exhibited a proactive deficit characterized by intact retention of the predelay information (i.e., arms entered prior to the delay) but impaired performance on the postdelay component (arms entered only after the delay). These data indicate: (i) that state dependency does not explain the working memory deficits induced by intraseptal CDP; (ii) that pretest CDP disrupts the storage and utilization of working memory for current arm selections. The second experiment examined the behavioral effects induced by a permanent disruption of the cholinergic septohippocampal pathway produced by icv injection of the cholinotoxin AF64A. Rats were initially trained on the DNMTS task and then bilaterally injected icv with either AF64A (2.5 nmoles/side) or CSF. AF64A-treated rats exhibited a significant impairment of performance compared to CSF-treated controls. In contrast to the impairment exhibited by CDP-treated rats in Experiment 1, the performance of AF64A-treated rats displayed a deficit in the maintenance/retrieval of information acquired during RAM training and an impairment in ability to store current spatial information in working memory to guide postdelay testing performance. These studies demonstrate that acute and chronic disruptions of the septohippocampal pathway produce distinct profiles of cognitive impairment that should help to reveal the behavioral and neurobiological characteristics of spatial memory.

Anatomical specificity and time-dependence of chlordiazepoxide-induced spatial memory impairments.

Injection of the benzodiazepine (BDZ) chlordiazepoxide (CDP) into the medial septum (MS) produced a dose-dependent retrograde working memory deficit in a delayed non-match-to-sample radial-arm maze task. CDP (30 nmol; 10 micrograms) decreased the number of correct choices and increased the number of errors without altering latency to make arm choices. The effects of CDP were site specific; injection into regions proximate to the MS, including the lateral septum, the anterior cingulate, and the nucleus basalis magnocellularis, did not affect any index of performance. The second experiment demonstrated that CDP impaired working memory only when rats were injected either 0 or 60 min, but not 15, 30, or 45 min, following training. The MS appears (a) to contribute to both early (encoding/ maintenance) and late (retrieval/utilization) phases of working memory and (b) to be a critical site of action for BDZ-induced deficits in spatial working memory.

A single intraseptal injection of nerve growth factor facilitates radial maze performance following damage to the medial septum in rats.

Rats were trained on a radial maze and then given electrolytic lesions of the MS followed by a single intraseptal injection of 5 micrograms of NGF. Three days later they were re-tested on the maze. They were also post-operatively tested for hyperemotionality. MS lesions severely impaired performance on the radial maze and produced increased emotionality. MS lesions also produced a general decrease in hippocampal high affinity choline transport and acetylcholinesterase staining, which was not affected by NGF administration. NGF treatment ameliorated the behavioral deficit in the radial maze but had no effect on the hyperemotionality. In order to determine whether the NGF was working to restore previously learned spatial abilities, the type of learning strategy used by the animals was also assessed. NGF treatment did not restore previously learned spatial strategies but facilitated recovery of alternative learning strategies. The reduction in cognitive deficit was also paralleled by reduced ventricular enlargement in the NGF treated rats. The present results suggest that a single injection of NGF can produce a long-lasting improvement on a cognitive task and reduce some of the injury-induced, secondary reactive changes that occur following electrolytic MS lesions.

Baclofen produces dose-related working memory impairments after intraseptal injection.

Altering the activity of the septohippocampal pathway can impair spatial memory in rats. Pharmacological manipulation of septal GABA-A receptors with the agonist, muscimol, or the benzodiazepine agonist, chlordiazepoxide, also impairs spatial memory and depresses hippocampal cholinergic activity. The present experiment examined the effects of intraseptal infusion of the GABA-B agonist baclofen on the performance of rats on a working memory radial arm maze (RAM) task. Post-training administration of baclofen (3 nmol, but not 1.5 or 0.75 nmol) produced a significant impairment of RAM performance. Baclofen significantly reduced the number of correct choices and increased the number of errors committed during testing without affecting latency per arm choice or the ability of the rats to navigate the maze and consume food pellets. The data suggest that baclofen impaired retention of the task without producing proactive performance deficits. Furthermore, the present data are consistent with the hypothesis that a GABAergic mechanism in the medial septum modulates the maintenance or retrieval of spatial working memory.

Strategies to limit brain injury and promote recovery of function.

Neurotoxicology integrates the best aspects of pharmacology, neuroscience, and toxicology and has established unique approaches, questions, and model systems. It has contributed to (i) isolating the multitude of environmental factors that damage the nervous system, (ii) characterizing the nature of the insult in both developing and mature organisms, and (iii) identifying the populations at risk. These efforts have symbolized the scope of modern neurotoxicology but there is now a growing appreciation that neurotoxicology, neurology, and biological psychiatry share common interests that are based upon a set of unified conceptual questions. The unifying goals for all of these disciplines are to discern the biology of neural insult, its functional consequences, and the extent and underlying mechanisms of recovery of function. While research is beginning to unravel the biology of neurological disorders there is a widening gap between our understanding of the substrates of these diseases and our ability to prevent or treat them. A concerted effort must be mounted to develop new and appropriate treatments for neurodegenerative disorders. The vulnerability of the hippocampus to a broad spectrum of insults (cerebrovascular insufficiency, excitotoxins, age-related neurodegenerative disorders, drugs of abuse) and its involvement in cognitive function makes it a logical focal point for the study of the behavioral and neurobiological correlates of recovery of function. This article discusses models of neurodegeneration and different therapeutic strategies that might limit injury or promote recovery of function.

Vitamin E prevents the place learning deficit and the cholinergic hypofunction induced by AF64A.

The present study examined whether pretreatment with vitamin E would attenuate the place learning impairment and the neurochemical deficits induced by intracerebroventricular (icv) injection of the cholinotoxin ethylcholine mustard aziridinium ion (AF64A). Male Sprague-Dawley rats were pretreated by intramuscular injection of either vitamin E or saline 24 h and 15 min prior to surgery. They were then infused bilaterally into the cerebroventricles with AF64A (3 nmol/side) or artificial cerebrospinal fluid (CSF). Following 14 days of postoperative recovery, the rats were trained to acquire a place learning task in a water maze for two blocks of four trials a day for 5 days. Following the 40th trial a probe trial was used to assess memory for the platform location. After completion of behavioral testing, hippocampal high-affinity choline uptake (HAChU) was assessed. The groups which had received either saline or vitamin E pretreatment followed by icv injection of CSF did not differ significantly on any parameter measured and were therefore pooled as control group. Animals which had been pretreated with vitamin E and had received icv injection of AF64A exhibited neither significant impairment in water maze performance nor significant decrease in HAChU. In contrast, animals which had been pretreated by saline followed by icv injection of AF64A were significantly impaired in acquisition of the place learning task as well as during the probe trial and had reduced HAChU in the hippocampus. These findings indicate that vitamin E may have a neuroprotective effect in the septohippocampal cholinergic system.

AF64A (ethylcholine mustard aziridinium) impairs acquisition and performance of a spatial, but not a cued water maze task: relation to cholinergic hypofunction.

The cholinergic neurotoxin AF64A (ethylcholine mustard aziridinium) produced alterations in a spatial but not a nonspatial cognitive task following ICV injection. AF64A impaired acquisition and performance in the standard Morris water maze task, evidenced by significantly longer latencies to find the submerged platform. However, the AF64A group exhibited shorter latencies and more direct paths to the target at the end of training, which suggests acquisition of efficient search strategies and a sparing of procedural memory. However, the AF64A group spent significantly less time in the target quadrant during the probe trial than the CSF group. This suggests a failure to learn the specific location of the target and impaired declarative memory processes. In contrast, AF64A did not affect performance of a cued MWM task that did not involve spatial memory processing, demonstrating the absence of motoric, sensory, or motivational impairments. The AF64A-induced behavioral deficits were associated with a) a significant decrease in high affinity choline transport (HAChT), b) reduced concentrations of 5-HT and 5-HIAA, and c) an increased ratio of 5-HIAA/5HT, in the HPC. There were no changes in choline uptake in the gustatory cortex, the amygdala, or the striatum. Percent time in the target quadrant during the probe trial was significantly correlated with HAChT in the HPC. There were no correlations between any of the behavioral measures and HAChT in the striatum, gustatory cortex, or the amygdala, or between serotonergic or noradrenergic parameters in the HPC. These data suggest that AF64A produces cognitive deficits in spatial tasks that are correlated with the cholinergic hypofunction induced by the compound.