Studies Debunk Hippocampus Memory Myths

Neuroscientists have often made the claim that the hippocampus is necessary for the formation of new memories. For example, one paper claimed that "clinical evidence indicates that damage to the hippocampus produces anterograde amnesia."  Anterograde amnesia is an inability to form new memories.  There was never any good evidence for such claims. 

To back up claims such as the one above, some people cite the case of patient H.M, a patient with a damaged hippocamous.  For example, the paper quoted above states that patient H.M. "became unable to consciously recollect new events in his life or new facts about the world."  This is not entirely correct. A 14-year follow-up study of patient H.M. (whose memory problems started in 1953) actually tells us that H.M. was able to form some new memories. The study says this on page 217:

"In February 1968, when shown the head on a Kennedy half-dollar, he said, correctly, that the person portrayed on the coin was President Kennedy. When asked him whether President Kennedy was dead or alive, and he answered, without hesitation, that Kennedy had been assassinated...In a similar way, he recalled various other public events, such as the death of Pope John (soon after the event), and recognized the name of one of the astronauts, but his performance in these respects was quite variable."

Another paper tells us that patient H.M. was able to learn new motor skills., stating this: "H.M. could successfully acquire, and subsequently retain, new motor skills in the context of several other experimental tasks (e.g., rotary pursuit, bimanual tracking, tapping)."  Another paper ("Evidence for Semantic Learning in Profound Amnesia: An Investigation With Patient H.M.") states this:

"We used cued recall and forced-choice recognition tasks to investigate whether the patient H.M. had acquired knowledge of people who became famous after the onset of his amnesia. Results revealed that, with first names provided as cues, he was able to recall the corresponding famous last name for 12 of 35 postoperatively famous personalities. This number nearly doubled when semantic cues were added, suggesting that his knowledge of the names was not limited to perceptual information, but was incorporated in a semantic network capable of supporting explicit recall. In forced-choice recognition, H.M. discriminated 87% of postmorbid famous names from foils. Critically, he was able to provide uniquely identifying semantic facts for one-third of these recognized names, describing John Glenn, for example, as 'the first rocketeer' and Lee Harvey Oswald as a man who 'assassinated the president.' Although H.M.’s semantic learning was clearly impaired, the results provide robust, unambiguous evidence that some new semantic learning can be supported by structures beyond the hippocampus proper."

It is not scientific to cite a patient with one physical issue and some other problem, and to claim or insinuate that the problem was caused by the physical issue. Using the same logic, you could take someone with hair loss and a problem concentrating, and claim that the problem concentrating was caused by the hair loss.  Ideas about a cause of something can only be soundly derived from studies involving many patients, not just one or a few. 

The main research paper on the hippocampus and memory is the paper "Memory Outcome after Selective Amygdalohippocampectomy: A Study in 140 Patients with Temporal Lobe Epilepsy." That paper gives memory scores for 140 patients who almost all had the hippocampus removed to stop seizures.  Using the term "en bloc" which means "in its entirety" and the term "resected" which means "cut out," the paper states, "The hippocampus and the parahippocampal gyrus were usually resected en bloc."  The paper refers us to another paper  describing the surgeries, and that paper tells us that hippocampectomy (surgical removal of the hippocampus) was performed in almost all of the patients. 

The "Memory Outcome after Selective Amygdalohippocampectomy" paper does not use the word "amnesia" to describe the results. That paper gives memory scores that merely show only a modest decline in memory performance.  The paper states, "Nonverbal memory performance is slightly impaired preoperatively in both groups, with no apparent worsening attributable to surgery."  In fact, Table 3 of the paper informs us that a lack of any significant change in memory performance after removal of the hippocampus was far more common than a decline in memory performance, and that a substantial number of the patients improved their memory performance after their hippocampus was removed. 

A 2020 paper is entitled "Preserved visual memory and relational cognition performance in monkeys with selective hippocampal lesions." The paper states this:

"We tested rhesus monkeys on a battery of cognitive tasks including transitive inference, temporal order memory, shape recall, source memory, and image recognition. Contrary to predictions, we observed no robust impairments in memory or relational cognition either within- or between-groups following hippocampal damage."

Citing a previous study, the paper notes that "formation of new memories in the object-in-scene task, one of the most accepted tests of episodic memory used with nonhuman primates, was found to be unaffected by lesions of the hippocampus itself."  It also notes that "There is a concerning lack of clear causal evidence for a critical role of the hippocampus in visual memory, episodic memory, recollection, or relational cognition in nonhuman primates."

To test the effects of hippocampus damage, the study authors injected five rhesus monkeys with neurotoxins. They estimate that this damaged about 75% of the hippocampus structures of the monkeys (Figure 1). The monkeys were subjected to a wide variety of cognitive tests. The paper concludes this:

"Contrary to dominant theories, we found no evidence that selective hippocampal damage in rhesus monkeys produced disordered relational cognition or impaired visual memory. Across a substantial battery of cognitive tests, monkeys with hippocampal damage were as accurate as intact monkeys and we found no evidence that the two groups of monkeys solved the tasks in different ways."

These results were similar to those reported by the paper here, entitled "Nonnavigational spatial memory performance is unaffected by hippocampal damage in monkeys." The study tested five monkeys. The study states the following, noting that the monkey that performed the best on one memory test was in the group of hippocampus-damaged monkeys, not the control group of normal monkeys:

"Hippocampal damage did not reduce memory span or slow acquisition. Monkeys with hippocampal damage and control monkeys did not differ in the memory span they achieved during training (mean: HP = 4.4, C = 3.8; median = 4 for both groups; t8 = 1.09, p = .305). The monkey that progressed to the longest memory span (6) was in the hippocampal group (Table 1)."

The 1998 paper "Object Recognition and Location Memory in Monkeys with Excitotoxic Lesions of the Amygdala and Hippocampus" gave 11 monkeys "selective lesions of the amygdala and hippocampus made with the excitotoxin ibotenic acid." According to Table 1, the average hippocampus damage for seven of the monkeys was 73%. We read the following

"Postoperatively, monkeys with the combined amygdala and hippocampal lesions performed as well as intact controls at every stage of testing. The same monkeys also were unimpaired relative to controls on an analogous test of spatial memory, delayed nonmatching-tolocation. It is unlikely that unintended sparing of target structures can account for the lack of impairment; there was a significant positive correlation between the percentage of damage to the hippocampus and scores on portions of the recognition performance test, suggesting that, paradoxically, the greater the hippocampal damage, the better the recognition."

A 2019 paper describing experiments with rhesus monkeys is entitled "Nonnavigational spatial memory performance is unaffected by hippocampal damage in monkeys."  A 2023 paper did a meta-analysis of many studies testing memory on monkeys who had been given lesions of the hippocampus. The study was entitled "Reevaluating the role of the hippocampus in memory: A meta-analysis of neurotoxic lesion studies in nonhuman primates." Here is figure 5B from the paper:

The graph is exaggerating the differences, because it using a scale starting at 50% rather than 0%, which is a graph trick that makes small differences look twice as big. Even with the "make the differences look bigger" trick, we see nothing very impressive in regard to the hippocampus. With a short delay and a long delay, there is merely a minimal difference, with the hippocampus-damaged monkeys performing a few percent worse. With a medium delay, the hippocampus-damaged monkeys performed a little bit better. These results fail to back up claims that the hippocampus is crucial for memory. 

Here is graph 6A from the paper. The graph plots the amount of hippocampus damage on one axis, and the performance on the memory test on the other axis.  The graph tells no clear tale. In three of the studies, very good performance (90% or better) occurred despite very high damage of the hippocampus (75% or more ). In seven of the studies, good performance (85% or better) occurred despite 50% or greater hippocampus damage. Figure 3C of the paper shows that the studies involving hippocampus damage of 75% or greater involved about 15 animals per study, while the studies involving hippocampus damage of less than 70% used an average of only about 8 subjects. So we should be granting more weight here to the results shown in the upper right of the diagram below, results showing heavy hippocampus damage and little performance damage. 

The 1997 paper "Differential Effects of Early Hippocampal Pathology on Episodic and Semantic Memory" reported on three persons with severe hippocampus damage. We read, "Volumetric measurements derived from three-dimensional (3D) data sets showed that in each of the three patients, the hippocampi are abnormally small bilaterally, with volumes ranging from 43 to 61% of the mean value of normal individuals (Figs. 2 and 3A)."  We learn that "all three patients are not only competent in speech and language but have learned to read, write, and spell." We read this:

 "With regard to the acquisition of factual knowledge, which is another hallmark of semantic memory...all three patients obtained scores within the normal range (Table 2). A remarkable feature of Beth’s and Jon’s stores of semantic memories is that they were accumulated after these patients had incurred the damage to their hippocampi."

In a long footnote to Table 2, we get examples of the three patients answering questions based on quite complex writing in front of them, and answering some common knowledge questions, and the answers sound as good as you or I might give. The authors of this paper attempt to persuade us that the three patients suffered from damage to episodic memory. But they give no very strong evidence of such a thing, mainly mentioning that "none is well oriented in date and time, and they must frequently be reminded of regularly scheduled appointments and events, such as particular classes or extracurricular activities," and that "none can provide a reliable account of the day’s activities or reliably remember telephone conversations or messages, stories, television programs, visitors, holidays, and so on," leaving us in the dark about what exactly they mean by "reliably." Did they mean 100% correct, 95% correct, or 90% correct? We can't tell.  Overall, the paper is inconsistent with claims about the hippocampus being essential for memory. 

Postscript: Harvard scientist Karl Lashley did extensive experiments with animals, experiments involving removal or damage to different parts of the brain. In much of what he wrote, it is hard to disentangle the effect of hippocampus damage. But on page 92 of his book Brain mechanisms and intelligence; a quantitative study of injuries to the brain, we have a table that makes it pretty easy to check for how much of an effect damage to the hippocampus has on maze performance in animals who had been trained to run a maze before parts of their brain were removed. The column on the far right lists the type of lesion the animal had. A letter H stands for hippocampus, N stands for No Injury, F stands for Fornix, R stands for right, L stands for left, and the numbers 1, 2 and 3 stand for grade of injury from slight to severe.

Lashley states this:

"If we select all cases which made more than 25 errors in retention tests, we find that there is no area of [brain] destruction common to all. For example, cases 100, 107 and 111 all show very serious [maze performance] loss, making from 5 to 47 times as many errors in postoperative retention tests as the normal average for learning. Their lesions are compared in Figure 23, which shows no significant overlap between them."

These findings were contrary to the dogma that the hippocampus is crucial to memory. Below is what the table tells us about some of the cases. The results are inconsistent with claims that the hippocampus is crucial for memory. 

Case #	Total brain tissue loss (%)	Hippocampus damage?	Total training time, seconds (post-operative)	Errors	Trials	Comment
98	21.1	Medium damage on right and left hippocampus	310	34	15	Much better performance than in case 100, which had no hippocampus damage but similar brain loss damage
96	20.6	Medium damage on  left hippocampus	63	5	1	Excellent  performance, with hippocampus damage and one fifth brain loss
100	21.5	None	7539	768	75	Weak performance, but no hippocampus damage
107	25.4	None	11536	689	150	Weak performance, but no hippocampus damage
111	28.3	Medium damage on right and left hippocampus plus septum damage	2230	127	48	Much better performance than cases 107 and 100, despite medium-level hippocampus damage and more brain tissue loss.
114	31.1	Small damage on both right and left hippocampus. About one third of brain removed.	12	1	1	Very good memory performance despite heavy brain damage and some hippocampus damage
116	33.9	Severe damage on right and left hippocampus	2836	547	150	Weak performance with very bad hippocampus damage and one third of brain loss, but much better performance than cases 100 and 107 where there was no hippocampus damage and  less brain damage

Case 116 in the data above (described in the last row of the table above) debunks claims that the hippocampus is essential for memory. In that case an animal with severe damage to both the right and left hippocampus was able to learn, and learn better than some animals with no hippocampus damage and less brain damage. Case 114 in the data above defies all claims that memories are stored in brains, as it involves excellent memory performance in an animal with one third of the brain removed. 

In the paper "Why There Are Complementary Learning Systems in the Hippocampus and Neocortex: Insights From the Successes and Failures of Connectionist Models of Learning and Memory" you can read here, we are told this:

"Some kinds of learning appear to be completely unaffected by hippocampal system lesions.... Examples of forms of learning that are spared are gradually acquired skills that emerge over several sessions of practice, such as the skill of tracing a figure viewed in a mirror (B. Milner, 1966), reading mirror-reversed print (N. J. Cohen & Squire, 1980), or anticipating subsequent items in a sequence governed by a complex stochastic grammar (Cleeremans, 1993). Hippocampal patients also appear to be spared in their ability to learn the structure common to a set of items: They are as good as normals in judging whether particular test items come from the same prototype, or were generated by the same finite-state grammar, as the members of a previously studied list (Knowlton, Ramus, & Squire, 1992; Knowlton & Squire, 1993)....In animal studies, it is clear that some forms of classical or instrumental conditioning of responses to discrete salient cues are unaffected by hippocampal system damage (for reviews, see Barnes, 1988; O'Keefe & Nadel, 1978; Rudy & Sutherland, 1994)."

The paper here refers to humans with hippocampus lesions, and tells us "their acquisition of new skills appears to be completely intact."

The paper "Hippocampal Lesion Patterns in Acute Posterior Cerebral Artery Stroke" did memory tests on patients with lesions of the hippocampus. We have some memory tests on patients who had damage to the hippocampus because of a stroke infarct, who are referred to below as HI patients (hippocampal infarct patients).  The groups referred to are those with a right hippocampus infarct, and those with a left hippocampus infarct. We read this:

"In the MMSE, the patients reached a score of 24.30±3.91 (lying in the mildly impaired range), with no difference between groups, t(18)=1.33, P=0.202. In the Clock Drawing Test, the patients reached a score of 2.84±1.26 (at the border of the normal range), with no difference between groups, t(17)=0.51, P=0.618."  

In regard to results of a RBMT test of long-term verbal memory, we read this: "Compared to normative samples, the scores of patients with left HI were within the mildly impaired range, whereas the scores of patients with right HI were only slightly below the mean of the normative sample."  Overall, this paper supports the claim that the hippocampus is not some crucial component of memory. The people with hippocampus damage have done only slightly less than normal on memory tests.