David Gaffan

Scene-specific memory for objects: A model of episodic memory impairment in monkeys with fornix transection

A series of five experiments investigated the relationship between object memory and scene memory in normal and fornix-transected monkeys. An algorithm created formally defined background and objects on a large visual display; the disposition of some particular objects in particular places in a particular background constitutes a formally defined scene. The animals learned four types of discrimination problem: (1) object-in-place discrimination learning, in which the correct (rewarded) response was to a particular object that always occupied the same place in a particular unique background, (2) place discrimination learning, in which the correct response was to a particular place in a unique background, with no distinctive object at that place, (3) object discrimination learning in unique backgrounds, in which the correct response was to a particular object that could occupy one or the other of two possible places in a unique background, and (4) object discrimination learning in varying backgrounds, in which the correct response was to a particular object that could appear at any place in any background. The severest impairment produced by fornix transection was in object-in-place learning. Fornix transection did not impair object discrimination learning in varying backgrounds. The results from the other two types of learning task showed intermediate severity of impairment in the fornix-transected animals. The idea that fornix transection in the monkey impairs spatial memory but leaves object memory intact is thus shown to be an oversimplification. The impairments of object memory in the present experiments are analogous to the impairments of episodic memory seen in human amnesic patients.

Projections from inferior temporal cortex to prefrontal cortex via the uncinate fascicle in rhesus monkeys

SummaryIn five rhesus monkeys (Macaca, mulatta) we used anterograde and retrograde tracing techniques to investigate the projection from the inferior temporal cortex (area TE) to the prefrontal cortex as well as the course of the projecting fibers. The results showed that TE projects to both the inferior convexity and orbital surface of prefrontal cortex and that these projections course almost exclusively via the uncinate fascicle. Transection of the uncinate fascicle deprives the prefrontal cortex of virtually all input from TE, but leaves intact inputs from prestriate and parietal visual areas as well as the amygdala. Such transection also leaves intact many projections from TE to targets other than the prefrontal cortex, including the amygdala, ventral putamen, tail of the caudate nucleus, and pulvinar.

Preserved Recognition Memory for Small Sets, and Impaired Stimulus Identification for Large Sets, Following Rhinal Cortex Ablations in Monkeys

Seven cynomolgus monkeys (Macaca fascicularis) performed a series of tasks designed to assess their visual memory and their ability to identify visual stimuli. Preoperatively they were trained and tested in delayed and simultaneous matching‐to‐sample, both with a large stimulus set and with a small stimulus set; there were ∼500 million possible stimuli in the large set, which effectively means that stimuli were trial‐unique with this set, while in the small set there were only four stimuli, which appeared repeatedly in every session of training with the small set. Three of the monkeys then had the cortex within and adjacent to the rhinal sulcus removed bilaterally, while the other four served as an unoperated control group. Postoperatively, the animals with ablation of the rhinal cortex showed severe impairment in delayed matching‐to‐sample with the large set. With the large set they were also impaired, however, in matching‐to‐sample with no delay between sample and test (0 s delay) and in simultaneous matching‐to‐sample, in which the sample and the two choice patterns were simultaneously present for inspection. The impairment in simultaneous matching‐to‐sample was particularly clear when the task was made more difficult by reducing the physical discriminability of the trial‐unique stimuli. With the small set of four stimuli, the animals with rhinal cortex ablation were not significantly impaired in overall performance level in delayed matching‐to‐sample, though their level was on average below that of the normal control animals. The stimulus set was then further restricted, so that there were now only two stimuli used throughout; in this condition, the animals with rhinal cortex ablation performed delayed matching‐to‐sample without any suggestion of impairment, showing indistinguishable performance levels from those of the control animals over a range of forgetting intervals. Subsequently, the animals were trained in trial‐unique non‐matching‐to‐sample with 0 s delay, which required reversal of the matching‐to‐sample rule they had previously learned; animals with rhinal cortex ablation showed a clear impairment in this rule‐reversal learning. The final experimental task was a concurrent discrimination learning task in which 20 pairs of stimuli were presented once per session; the animals with rhinal cortex ablation learned more slowly than the control animals on average, but the difference between the groups did not attain statistical significance. Overall, this pattern of deficits and of preserved abilities is clearly inconsistent with the idea that rhinal cortex ablation produces an impairment in all forms of visual recognition memory, and only in visual recognition memory. Instead, the present results indicate a general impairment in the capacity for knowledge about visual stimuli.

Dynamic Coding for Cognitive Control in Prefrontal Cortex

Summary Cognitive flexibility is fundamental to adaptive intelligent behavior. Prefrontal cortex has long been associated with flexible cognitive function, but the neurophysiological principles that enable prefrontal cells to adapt their response properties according to context-dependent rules remain poorly understood. Here, we use time-resolved population-level neural pattern analyses to explore how context is encoded and maintained in primate prefrontal cortex and used in flexible decision making. We show that an instruction cue triggers a rapid series of state transitions before settling into a stable low-activity state. The postcue state is differentially tuned according to the current task-relevant rule. During decision making, the response to a choice stimulus is characterized by an initial stimulus-specific population response but evolves to different final decision-related states depending on the current rule. These results demonstrate how neural tuning profiles in prefrontal cortex adapt to accommodate changes in behavioral context. Highly flexible tuning could be mediated via short-term synaptic plasticity.

Dissociated effects of perirhinal cortex ablation, fornix transection and amygdalectomy : evidence for multiple memory systems in the primate temporal lobe

Four experiments were performed with macaque monkeys (rhesus, Macaca mulatta, and cynomolgus, M.fascicularis). In experiment 1 six rhesus monkeys learned pre-operatively to perform delayed matching-to-sample, with complex naturalistic scenes as the stimulus material. Three of these monkeys then received bilateral ablations of the perirhinal cortex, while the other three received fornix transection. Both groups showed an impairment postoperatively, but the effect of perirhinal cortex ablation was significantly more severe than the effect of fornix transection. In experiment 2 the same animals, together with three normal, control rhesus monkeys,which had a similar training history, performed simple, spatial discrimination learning in a Wisconsin General Test Apparatus. The animals with fornix transection were impaired, but the animals with ablations of perirhinal cortex were not. In experiment 3 the nine animals from experiment 2 were tested for the acquisition of systematic preferences among four novel foods (apple, lemon, olive, meat). Their results were compared with those from a previously published experiment with normal and amygdalectomized cynomolgus monkeys which had been given the same food preference test. Amygdalectomy produced a significant disruption of food preference learning but the other two lesions (fornix transection and perirhinal cortex ablation) did not. In experiment 4, 16 rhesus monkeys (9 normal controls, 4 with perirhinal cortex ablation, and 3 with fornix transection) learned to discriminate among complex naturalistic scenes, in a task in which each scene was presented only once per day in the main part of the experiment. The two operated groups were impaired, and there was no significant difference between the severity of the impairments. Thus, the effects of perirhinal cortex ablation can be doubly dissociated from the effects of fornix transection (experiments 1 and 2) and both can be dissociated from the effects of amygdalectomy (experiment 3). Furthermore, the results of experiment 4 show that the effects of perirhinal cortex ablation are not limited to tasks of memory over short retention intervals. On the basis of the presently reported data and other known effects of perirhinal cortex ablation, it is suggested that this ablation produces an impairment in knowledge (semantic memory) about objects.

Monkeys (Macaca fascicularis) with rhinal cortex ablations succeed in object discrimination learning despite 24-hr intertrial intervals and fail at matching to sample despite double sample presentations.

Six cynomolgus monkeys (Macaca fascicularis) learned preoperatively a set of 10 concurrent object discriminations with 24-hr intertrial intervals. Three then had the rhinal cortex removed bilaterally, whereas the other 3 remained as unoperated controls. The animals with ablations were impaired in reacquiring the preoperatively acquired set but subsequently learned without any impairment a new set of 10 discriminations that was presented in the same way. The monkeys with rhinal cortex ablations then failed to learn delayed matching to sample, with double sample presentations, in 510 trials, whereas the control animals learned this task in 270 trials on average. The results add to existing evidence that rhinal cortex ablation produces a severe impairment in visual short-term recognition memory and show for the first time that this impairment is accompanied by normal long-term discrimination learning ability.

Loss of recognition memory in rats with lesions of the fornix

Abstract Anterograde amnesia, which is caused in man by lesions of the hippocampo-mammillary system, may fundamentally consist in a deficit of recognition memory (familiarity discrimination). Six experiments on the behavioural effects of similar lesions in rats demonstrated an inability to discriminate familiarity in the lesioned animals and suggested that this deficit may explain the pattern of results in previous experiments on hippocampal lesions in animals.

Place memory and scene memory: effects of fornix transection in the monkey

SummaryFive experiments examined the effects of fornix transection upon some spatial and visual learning tasks in monkeys (Macaca fascicularis). For each trial of each task, the monkey was brought to a test tray and allowed to choose between 2 objects on the tray. In different tasks, different cues were provided by the experimenter to guide the monkeys choices. In total 5 different tasks were run (Experiments 1 to 5) and the results showed that the effects of fornix transection varied markedly between tasks: the animals with fornix transection were severely impaired in experiments 1, 3 and 5 but learned normally in experiments 2 and 4. It is concluded that the results cannot be explained by the simple hypothesis of a deficit in place learning, since some forms of place learning are unimpaired by fornix transection. A better general hypothesis is that the memory disrupted by fornix transection is like a snapshot memory, which stores the spatial arrangement of items in a witnessed scene.

Filtering of neural signals by focused attention in the monkey prefrontal cortex.

Prefrontal cortex is thought to be important in attention and awareness. Here we recorded the activity of prefrontal neurons in monkeys carrying out a focused attention task. Having directed attention to one location, monkeys monitored a stream of visual objects, awaiting a predefined target. Although neurons rarely discriminated between one non-target and another, they commonly discriminated between targets and non-targets. From the onset of the visual response, this target/non-target discrimination was effectively eliminated when the same objects appeared at an unattended location in the opposite visual hemifield. The results show that, in prefrontal cortex, filtering of ignored locations is strong, early and spatially global. Such filtering may be important in blindness to unattended signals—a conspicuous aspect of human selective attention.

A comparison of the effects of fornix transection and sulcus principalis ablation upon spatial learning by monkeys

In each of three experiments with Cynomolgus monkeys (Macaca fascicularis), there was a group of normal control animals, a group with bilateral cortical ablations in the principal sulcus, and a group with fornix transection. In Expt. 1, half of each group learned problems in which the position of a pair of visual stimuli, to the monkeys left or right, indicated which of the visual stimuli was the correct (rewarded) one. The other animals learned problems in which visual stimuli indicated, irrespective of their own spatial position, whether reward was to be found on the monkeys left or on the right. The animals with fornix transection were impaired in both tasks. The animals with sulcus principalis ablation were also impaired in both tasks. The impairment caused by fornix transection was more severe than that caused by sulcus principalis ablation. Within each of the two operated groups, the degree of impairment in the two tasks was equal, when assessed in proportion to the difficulty of each task for control animals. Expt. 2 showed that neither of the operated groups was impaired in visual discrimination learning with spatial position irrelevant. Expt. 3 tested spatial discrimination learning (acquisition and reversal of a left-right discrimination) with irrelevant visual cues. Here the fornix-transected group was impaired but the group with sulcus principalis ablations was normal. It is suggested, on the basis of these findings and previous results, that fornix transection produces a general deficit in remembering the spatial arrangement of whole scenes, while sulcus principalis ablation produces a deficit in high-order integration involving spatial information.