Excitotoxic Hippocampal Lesions Research Articles

Cell-type specific circadian transcription factor BMAL1 roles in excitotoxic hippocampal lesions to enhance neurogenesis

Circadian clocks, generating daily rhythms in biological processes, maintain homeostasis in physiology, so clock alterations are considered detrimental. Studies in brain pathology support this by reporting abnormal circadian phenotypes in patients, but restoring the abnormalities by light therapy shows no dramatic effects. Recent studies on glial clocks report the complex effects of altered clocks by showing their beneficial effects on brain repairs. However, how neuronal clocks respond to brain pathology is elusive. This study shows that neuronal BMAL1, a core of circadian clocks, reduces its expression levels in neurodegenerative excitotoxicity. In the dentate gyrus of excitotoxic hippocampal lesions, reduced BMAL1 in granule cells precedes apoptosis. This subsequently reduces BMAL1 levels in neighbor neural stem cells and progenitors in the subgranular zone, enhancing proliferation. This shows the various BMAL1 roles depending on cell types, and its alterations can benefit brain repair. Thus, cell-type-specific BMAL1 targeting is necessary to treat brain pathology.

In the temporal organization of episodic memory, the hippocampus supports the experience of elapsed time.

Space and time are both essential features of episodic memory, for which the hippocampus is critical (Howard & Eichenbaum, 2015). Spatial tasks have been used effectively to study the behavioral relevance of place cells. However, the behavioral paradigms utilized for the study of time cells have not used time duration as a variable that animals need to be aware of to solve the task. Therefore, the behavioral relevance of this cell firing is unclear. In order to directly study the role of the hippocampus in processing elapsed time, we created a novel time duration discrimination task. Rats learned to make a decision to turn left or right depending on the preceding tone duration (10 s, left turn; 20 s, right turn). Once the rats reached criterion performance of 90% correct on two out of three consecutive days, they received either an excitotoxic hippocampal lesion or a sham-lesion surgery. After recovery, rats were tested to determine hippocampal involvement in discriminating time duration. Rats with hippocampal lesions performed at chance level on their first testing day postlesion, and they were impaired relative to the sham-lesioned rats. Although the hippocampal-lesioned rats began discriminating at above chance level, their performance never returned to criterion even with 50 days of postoperative testing. Furthermore, while sham rats showed no difference in the number of errors they made on 10- versus 20-s delay trials, hippocampal lesion rats similarly improved their performance under the 10-s delay condition, but not under the 20-s delay condition. Results indicate that hippocampal lesions resulted in a selective impairment in discriminating elapsed time only during the longer delay trials. The implications of these results are discussed in relation to the limits of working-memory capacity and to the role of sustained hippocampal time cell activity in memory performance depending on the perceived relevance of the delay period.

Latent inhibition of conditioned taste aversion in rats with excitotoxic dorsal hippocampal lesions.

The hippocampus plays crucial roles for the acquisition of latent inhibition in different associative learning procedures, such as fear conditioning. However, the involvement of the hippocampus in the latent inhibition of conditioned taste aversion (CTA) is uncertain. Because different subregions of the hippocampus are associated with distinct functions, it is possible that specific regions of this structure are selectively involved in this learning. To explore the relationship between the dorsal hippocampal region and the latent inhibition of CTA, we analyzed the behavioral effects of excitotoxic lesions of the dorsal hippocampus vs. sham lesions in this paradigm. The results provide no evidence that the latent inhibition of CTA is compromised in rats with excitotoxic dorsal hippocampal lesions. The differential involvement of specific hippocampal regions in the latent inhibition of other associative learning paradigms is briefly discussed.

Procedural Performance Benefits after Excitotoxic Hippocampal Lesions in the Rat Sequential Reaction Time Task.

It is widely agreed upon that hippocampal function is linked to episodic-like and spatial memory across various species, for example, rodents. However, the interplay between hippocampal function and other types of learning and memory, like procedural stimulus-response or sequential learning, is less clear. Recently (Eckart et al. in Hippocampus 22:1202-1214, 2012), we showed that excitotoxic hippocampal lesions, which mainly affected its dorsal part, led not only to the expected deficits in a spatial and episodic-like memory task, namely the object place recognition test, but also to substantial improvements in terms of speed and accuracy in a rat adaption of the human sequential reaction time task (SRTT). The design of that experiment, however, which included fixed test durations per training day, led to the fact that lesioned animals gained more instrumental experience, which may partly have accounted for their enhanced performance. In order to rule out such a potential confound, we performed the present experiment on rats with similar ibotenic lesions aiming at the dorsal hippocampus, but we now kept the amount of correct instrumental responses and reinforcements on the same level as in controls. Our data revealed that lesioned animals were still able to complete the SRTT in a substantially smaller amount of time, when compared to control and sham-operated animals, although no differences were observable in terms of speed or accuracy. Also, the animals with lesions showed impaired extinction in a subsequent test where rewards were omitted. The former effect can primarily be attributed to shorter post-reinforcement pauses in the lesioned animals, and the possible mechanisms of this and the extinction effect will be addressed in the discussion.

Changing the rate and hippocampal dependence of trace eyeblink conditioning: Slow learning enhances survival of new neurons

Trace eyeblink conditioning in which a conditioned stimulus and unconditioned stimulus are separated by a gap, is hippocampal dependent and can rescue new neurons in the adult dentate gyrus from death (e.g., Beylin et al., 2001; Gould et al., 1999). Tasks requiring more training trials for reliable expression of the conditioned response are most effective in enhancing survival of neurons ( Waddell & Shors, 2008). To dissociate hippocampal dependence from acquisition rate, we facilitated hippocampal-dependent trace eyeblink conditioning in two ways: a shorter trace interval and signaling the intertrial interval with a post-US cue. Trace conditioning with a shorter trace interval (250 ms) requires an intact hippocampus, and acquisition is faster relative to rats trained with a 500 ms trace interval (e.g., Weiss et al., 1999). Using excitotoxic hippocampal lesions, we confirmed that eyeblink conditioning with the 250 or 500 ms trace interval is hippocampal dependent. However, training with the post-US cue was not hippocampal dependent. The majority of lesion rats in this condition reached criterion of conditioned responding. To determine whether hippocampal dependence is sufficient to rescue adult-generated neurons in the dentate gyrus, rats were injected with BrdU and trained in one of the three trace eyeblink arrangements one week later. Of these training procedures, only the 500 ms trace interval enhanced survival of new cells; acquisition of this task proceeded slowly relative to the 250 ms and post-US cue conditions. These data demonstrate that rate of acquisition and not hippocampal dependence determines the impact of learning on adult neurogenesis.

Distinct contributions of the amygdala and hippocampus to fear expression.

The present study attempted to distinguish the independent contributions of the amygdala and hippocampus to fear expression. Rhesus monkeys (Macaca mulatta) with bilateral excitotoxic amygdala lesions (n = 4), bilateral excitotoxic hippocampal lesions (n = 8) and unoperated controls (n = 9) were allowed to reach over a neutral junk object or fear-provoking stimulus (i.e., a rubber snake or a jumping rubber spider) to retrieve a food reward. Monkeys were exposed to each stimulus for 30 s. On each trial we recorded the monkey's latency to retrieve the food reward and scored their whole-body reactions to the object. Confirming previous work we found that, relative to controls, both operated groups showed shorter food-retrieval latencies and exhibited fewer defensive and more approach behaviors when exposed to the fear-provoking stimuli. However, only monkeys with amygdala lesions showed an abnormal, excessive visual interest in the snake and spider. By contrast, monkeys with hippocampal lesions displayed behaviors that were unrelated to the presence of the fear stimuli, thereby indicating a lack of interest in, and emotional reactivity towards, the snake and spider. These data show that the hippocampus and amygdala contribute independently to the overall expression of defensive responses.

Dorsal, ventral, and complete excitotoxic lesions of the hippocampus in rats failed to impair appetitive trace conditioning

Three experiments examined appetitive trace and delay conditioning of the licking response (LR). In Experiment 1, normal rats were trained in trace conditioning using different trace intervals (2, 4, or 8 s) and in delay conditioning (i.e., with a 0-s trace) in order to determine an appropriate trace interval for the following lesion experiments. Only the rats trained with a 2-s trace interval ultimately reached the same level of learning as rats trained in delay conditioning. In Experiments 2A and 2B, the performance of rats with dorsal, ventral, and complete excitotoxic hippocampal lesions was compared to that of sham-operated rats in LR conditioning with a 2-s trace. In Experiment 2B, the performance of rats in trace LR conditioning was also compared to that of rats tested in the delay paradigm. In both experiments, acquisition did not differ in lesioned and sham-operated rats and, in Experiment 2B, it was faster in the delay than in the trace paradigm. These results contrast with those showing that aversive trace conditioning is impaired after hippocampal damage. Experiment 3 examined whether the differential effects of hippocampal lesions on aversive and appetitive trace conditioning could be related to a parametric difference, that is, the relative durations of the conditional stimulus and of the trace interval. Again, hippocampal damage failed to produce a learning impairment. It is suggested that the procedure of aversive, but not of appetitive, trace conditioning is context-specific and that an intact hippocampus is required only in these situations.

Processing idiothetic cues to remember visited locations: Hippocampal and vestibular contributions to radial‐arm maze performance

This research examined whether rats can use idiothetic cues to form spatial memories in the radial-arm maze (RM) and whether the hippocampus is involved in such ability. A possible contribution of the vestibular system to RM performance was also investigated. Rats with excitotoxic hippocampal lesions and sham-operated controls were trained on two versions of the RM task. In the Light condition, a unique visual insert was apposed on each arm floor and rats could choose which arm to enter next by relying on visual and/or idiothetic stimuli. In the Dark condition, the task was administered in darkness and success required processing of idiothetic cues to remember visited locations on the maze. In experiment 1, the performance of lesioned rats was impaired in the Light condition, but both control and lesioned rats learned to avoid already visited arms. In the Dark condition, the performance of controls improved over time whereas a severe deficit was observed in rats with hippocampal lesions. Thus, control rats, but not hippocampal lesioned rats, can form spatial memories by processing idiothetic inputs. Experiment 2 showed that vestibular lesions disrupt performance in both the Light and the Dark conditions and confirmed that rats use idiothetic information, especially vestibular cues, while navigating in the RM. Therefore, cues generated during locomotion play an important role in hippocampal-dependent spatial memory.

Context- but not familiarity-dependent forms of object recognition are impaired following excitotoxic hippocampal lesions in rats.

Dual-process models of recognition memory in animals propose that recognition memory is supported by two independent processes that reflect the operation of distinct brain structures: a familiarity process that operates independently of the hippocampus and a context-dependent (episodic) memory process that is dependent on the hippocampus. A novel variant of an object recognition procedure was used to examine this proposal. Healthy rats showed a preference for exploring a novel object rather than a familiar object: a familiarity-dependent recognition effect. They also showed a preference for exploring a familiar object that was presented in a different spatiotemporal context rather than a familiar object that was presented either in a different spatial or temporal context: a context-dependent form of recognition that is sensitive to "what" object has been presented "where" and "when." Rats with excitotoxic hippocampal lesions showed the familiarity-dependent but not the context-dependent form of recognition. The results provide support for dual-process theories of recognition memory.

GluR-A-dependent synaptic plasticity is required for the temporal encoding of nonspatial information.

Four related experiments studied operant performance of mice on differential reinforcement of low rates of responding (DRL) paradigms. Experiment 1 showed that excitotoxic hippocampal lesions impaired performance of a 10-s DRL schedule (DRL-10). Experiments 2 and 3 showed that GluR-A AMPA receptor subunit knockout mice, which are deficient in CA3-CA1 long-term potentiation (LTP), were markedly impaired at 15 s (DRL-15), but less impaired at DRL-10. Experiment 4 compared DRL-15 performance in mice from the 2 strains from which the GluR-A colony was derived and showed that they did not differ. The results show that GluR-A-containing AMPA receptors are required for normal performance on hippocampus-dependent, nonspatial working memory tasks, consistent with a role for GluR-A in the temporal encoding (what happened when) of nonspatial information.

The hippocampus and appetitive Pavlovian conditioning: Effects of excitotoxic hippocampal lesions on conditioned locomotor activity and autoshaping

The hippocampus (HPC) is known to be critically involved in the formation of associations between contextual/spatial stimuli and behaviorally significant events, playing a pivotal role in learning and memory. However, increasing evidence indicates that the HPC is also essential for more basic motivational processes. The amygdala, by contrast, is important for learning about the motivational significance of discrete cues. This study investigated the effects of excitotoxic lesions of the rat HPC and the basolateral amygdala (BLA) on the acquisition of a number of appetitive behaviors known to be dependent on the formation of Pavlovian associations between a reward (food) and discrete stimuli or contexts: (1) conditioned/anticipatory locomotor activity to food delivered in a specific context and (2) autoshaping, where rats learn to show conditioned discriminated approach to a discrete visual CS+. While BLA lesions had minimal effects on conditioned locomotor activity, hippocampal lesions facilitated the development of both conditioned activity to food and autoshaping behavior, suggesting that hippocampal lesions may have increased the incentive motivational properties of food and associated conditioned stimuli, consistent with the hypothesis that the HPC is involved in inhibitory processes in appetitive conditioning.

Hippocampal lesions, species-typical behaviours and anxiety in mice

The hippocampus is believed to play an important role in spatial cognition and anxiety. Much of the supporting evidence is derived from rat studies. Recent reports on hippocampal lesioned mice also showed impairments in spatial function, but anxiety was not uniformly diminished. There were, however, striking impairments in several “species typical” behaviours; lesioned mice made poorer nests, and hoarded and burrowed less. In the present experiments, mice with excitotoxic hippocampal lesions were tested in a well-established anxiety paradigm, the light–dark box. As in previous anxiety tests, the results were mixed; some measures (reduced dark time) suggested lesioned mice were less anxious; others (fewer light–dark transits) suggested greater anxiety. However, lesioned mice only made fewer transits when the door was small. This suggested that the tendency to enter small holes, so characteristic of small rodents, was reduced; subsequent tests showed lesioned mice preferred to explore in an alley rather than enter its attached tunnels. Further tests of “species typical” behaviours revealed that lesioned mice spent less time digging and climbing, and made less use of cardboard shelters in their cages. This was not due to inactivity; lesions did not reduce grooming or locomotion. Finally, tests of hyponeophagia showed hippocampal lesions reduced this measure of anxiety, so long as the control baseline was sufficiently high. Overall, the results suggest that the hippocampus is important in many species-typical behaviours, potentially influencing performance in a range of behavioural tests. However, species-typical behaviours offer easy and economical ways to test for hippocampal dysfunction, for example, in genetically modified mice.

Hippocampal lesions modulate both associative and nonassociative priming.

In associative priming, rats are more likely to orient to a visual stimulus whose memory has not been recently activated (V1) than to one whose memory has been recently activated (V2). However, rats with excitotoxic hippocampal lesions are more likely to orient to the primed V2 than to the unprimed V1. This study investigated the influence of hippocampal lesions on nonassociative priming. Rats received presentations of 2 visual stimuli, V1 and V2, that had been presented more (V2, primed) or less (V1, unprimed) recently. Control rats oriented to V1 rather than to V2, whereas hippocampal rats oriented to V2 rather than to V1. These results parallel those observed in an associative priming procedure and thereby suggest that the role of the hippocampus in priming is general.

Dissociation of function within the hippocampus: effects of dorsal, ventral and complete excitotoxic hippocampal lesions on spatial navigation

The present study was designed to assess the possibility that sub-total ventral hippocampal lesions might leave intact a mechanism for only highly accurate navigation, whereas sub-total dorsal hippocampal lesions might leave intact a mechanism only for less precise navigation. Animals with selective dorsal, ventral or complete hippocampal lesions were tested in a water maze, in which the target platform was moved from trial to trial, but always within a defined area, instead of being at a fixed location. Hence, an animal that searched at exactly the point where the platform had been found on a previous trial would be disadvantaged, in comparison with an animal that searched in the right general area. This might favor animals capable of less precise navigation over those with very precise navigational abilities. In subsequent phases of the experiment, we additionally assessed, for comparison, performance with a fixed platform location, reversal learning in the water-maze, and performance on an elevated T-maze. Our results revealed no sign of any qualitative difference between the effects of the selective sub-total lesions when the water maze hidden platform location was varied within the defined area, and the effects in subsequent more conventionally used tests. Ventral hippocampal damage never led to a performance deficit. Dorsal hippocampal damage led to significantly poorer performance in only some test phases, and never led to any sign of improved performance.

NMDA receptor blockade and hippocampal neuronal loss impair fear conditioning and position habit reversal in C57Bl/6 mice

The interpretation of learning and memory deficits in transgenic mice has largely involved theories of NMDA receptor and/or hippocampal function. However, there is little empirical data that describes what NMDA receptors or the hippocampus do in mice. This research assessed the effects of different doses of the NMDA receptor antagonist, MK-801, or different-sized hippocampal lesions on several behavioral parameters in adult male C57Bl/6 mice. In the first set of experiments, different doses of MK-801 (0.05–0.3 mg/kg, s.c.) were assayed in fear conditioning, shock sensitivity, locomotion, anxiety, and position habit reversal tests. Contextual and cued fear conditioning, and position habit reversal were impaired in a dose-dependent manner. Locomotor activity was increased immediately after injection of the highest dose of MK-801. A second set of experiments determined the behavioral effects of a moderate and large excitotoxic hippocampal lesion. Both lesions impaired contextual conditioning, while the larger lesion interfered with cued conditioning. Reversal learning was significantly diminished by the large lesion, while the moderate lesion had a detrimental effect at a trend level ( P<0.10). These results provide important reference data for studies involving genetic manipulations of NMDA receptor or hippocampal function in mice. Furthermore, they serve as a basis for a non-transgenic mouse model of the NMDA receptor or hippocampal dysfunction hypothesized to occur in human cognitive disorders.

Glycogen synthase kinase (GSK)-3β levels and activity in a neurodevelopmental rat model of schizophrenia

We have previously reported reduced GSK-3β protein levels and GSK-3 total (α+β isoforms) activity in postmortem frontal cortex of schizophrenic patients. We now studied whether GSK-3β is altered in the frontal cortex of rats with the neonatal excitotoxic hippocampal lesion used as a model of schizophrenia. Rats were infused with ibotenic acid (or artificial CSF in controls) bilaterally into the ventral hippocampus (VH) at postnatal day 7, then killed at postnatal day 35 (pre-puberty) or 56 (post-puberty). GSK-3β protein levels were reduced in the frontal cortex of the lesioned rats as compared to sham animals; post-hoc comparisons revealed that the reduction was statistically significant at a pre-pubertal age. Total GSK-3 (α+β) activity was not different between lesioned and sham rats at any age. These results demonstrate that reduced frontal cortical GSK-3β levels may occur as a result of neonatal hippocampal damage and suggest that this animal model may be utilized to study the mechanism of GSK-3 reduction in schizophrenia, a disorder in which postmortem changes in GSK-3 were found.

Unilateral hippocampal and inferotemporal cortex lesions in opposite hemispheres impair learning of single-pair visual discriminations as well as visuovisual conditional tasks in monkeys

Monkeys with unilateral ablations of the inferotemporal (IT) cortex were not impaired on learning or retention of single-pair object discriminations or visuovisual conditional tasks. Addition of an excitotoxic hippocampal lesion to the hemisphere opposite to the IT ablation impaired retention and acquisition of single-pair object discriminations and visuovisual conditional tasks. Histology revealed no areas of bilaterally symmetrical damage. Previous experiments have shown that bilateral excitotoxic hippocampal lesions do not impair single-pair object discriminations although they do produce a substantial impairment on visuovisual conditional tasks. Bilateral IT ablations produce impairment on single-pair object discrimination tasks. It is argued that the hippocampus in the hemisphere with the IT ablation is deprived of feed-forward visual input and that this, in addition to the contralateral hippocampal lesion, accounts for the impairment on the visuovisual conditional tasks. It is also argued that feed-back projections from the hippocampus to the IT cortex influence the learning of single-pair object discriminations. This influence may be difficult to demonstrate by the addition of hippocampal lesions to IT lesions because of the substantial effect of the IT lesion alone. It may be difficult to demonstrate by bilateral hippocampal lesions alone since the effect may be below that which generates an observable impairment. Nonetheless, an effect may be seen when a hippocampal lesion is made in monkeys with some IT damage, as in this experiment, as well as by the general observation that large lesions of the temporal lobes produce larger perceptuo-mnemonic impairments than lesions confined to the hippocampus or temporal neocortex in monkeys and man.

The Effects of Excitotoxic Hippocampal Lesions in Rats on Risperidone- and Olanzapine-Induced Locomotor Suppression

Previous studies have shown that excitotoxic hippocampal lesions in rats attenuate the ability of different doses of haloperidol, but not of clozapine, to suppress locomotor activity. The purpose of the present study was to determine if kainic acid-induced hippocampal damage reduces the degree of locomotor suppression produced by two relatively newer antipsychotic drugs, risperidone and olanzapine. Young adult male rats received bilateral intracerebroventricular infusions of the excitotoxin, kainic acid (KA), or vehicle and were tested for locomotor responses to drug treatment 30 days later. Infusions of KA produced neuronal loss in the CA3 region of the dorsal hippocampus in every rat. As reported previously, KA lesions reduced the ability of haloperidol (0.35 mg/kg) to completely suppress the locomotor activity elicited by amphetamine (1.5 mg/kg) relative to the effect of haloperidol in non-lesioned controls. Lesioned animals treated with a moderate dose of risperidone (1.4 mg/kg) also exhibited significantly more locomotor activity after amphetamine treatment in comparison to control animals. A trend toward greater activity was also observed in the lesioned group relative to the control group after olanzapine (3.0 mg/kg) injection (p = .09, 2-tailed). The locomotor effects of lower and higher doses of risperidone and olanzapine were not altered by kainic acid lesions. These data suggest that the locomotor-suppressive effects of moderate doses of risperidone and, perhaps, olanzapine involve hippocampal neurons, but that higher doses of each drug can suppress activity in a hippocampal-independent manner.

Transient expression of the NG2 proteoglycan by a subpopulation of activated macrophages in an excitotoxic hippocampal lesion.

Cells that express the NG2 proteoglycan (NG2+ cells) constitute a large glial population in the normal mature rodent brain. They can differentiate into oligodendrocytes but are distinct from mature oligodendrocytes, astrocytes, microglia, and neurons. Changes in NG2+ cells were examined in kainic acid-induced excitotoxic lesions of the hippocampus, and the relationship between NG2+ cells and reactive astrocytes and microglia was investigated between 1 and 90 days after lesioning. Two types of reactive NG2+ cells with altered morphology and increased NG2 immunoreactivity were observed in the lesion. Early changes, consisting of an increase in NG2 immunoreactivity and the number of processes, were apparent 24 h after lesioning and persisted through 3 months. These cells were distinct from reactive astrocytes or activated microglia/macrophages. A second type of reactive NG2+ cells appeared 2 weeks after injection, following an influx of macrophages. They had large, round cell bodies with short processes and expressed the microglia/macrophage antigens OX42 and ED1. Single cells coexpressing NG2 and macrophage/microglial antigens could be isolated from the lesion. The number of NG2+/OX42+ cells gradually declined and disappeared by 3 months after injection. They did not express glial fibrillary acidic protein or the alpha receptor for platelet-derived growth factor, indicating that they are distinct from astrocytes or oligodendrocyte progenitor cells. Cells that coexpressed NG2 and OX42 were never observed in hippocampal slice cultures treated with kainic acid, suggesting that NG2+/OX42+ cells are not derived from endogenous resident brain cells. These findings demonstrate that NG2 expression is transiently upregulated on activated macrophages/microglia that appear during the chronic stage in an excitotoxic lesion in the adult CNS.

Visual agnosia and Klüver–Bucy syndrome in marmosets ( Callithrix jacchus) following ablation of inferotemporal cortex, with additional mnemonic effects of immunotoxic lesions of cholinergic projections to medial temporal areas

Inferotemporal ablations in the New World monkey, the common marmoset ( Callithrix jacchus), produced a persistent impairment on visual discrimination learning and a florid, but transient, Klüver–Bucy syndrome. Monkeys with these ablations were impaired on acquisition of object discriminations to a high criterion and on concurrent discrimination learning, to a single high criterion across all trials. Neither the control monkeys nor the monkeys with inferotemporal ablations found acquisition more difficult when the component discriminations of a set were presented concurrently compared to consecutively, although the monkeys with inferotemporal ablations found acquisition under both these conditions somewhat more difficult than did control monkeys. This suggests that the severe impairment caused by inferotemporal ablations on concurrent learning measured across all trials is due to the need for sustained performance across a concurrent set rather than to the extra mnemonic demands of concurrent presentation. When immunotoxic lesions of the cholinergic projection to the hippocampal formation were added to the inferotemporal ablations, a further impairment on retention, and a differential impairment on concurrent, compared to consecutive, learning was observed. Previous studies have shown that lesions of the cholinergic projection to the hippocampus alone, or excitotoxic hippocampal lesions, do not affect simple visual discrimination learning. It is suggested that large inferotemporal ablations in monkeys produce a visual agnosia which causes severe ‘psychic blindness’ in the first instance, and a persistent impairment on visual discrimination learning. The hippocampus makes a contribution, which may be mnemonic, to discrimination performance after inferotemporal ablations.