lf.A personal publicationItaliano
Beyondliminalfinds.com

Above Average, Yet Far Below His Own Baseline

Five years after brain aneurysm surgery, a former programmer scored 115 on general memory and 110 on processing speed—both classified as high average—while 3D MRI showed cavitated lesions in both columns of the fornix.

Graphite pencil drawing on pale paper of a tall stack of bound notebooks standing on a bare surface beside a single closed notebook lying flat; two thin notebooks are missing from the middle of the stack, leaving a shadowed horizontal gap while the upper stack remains taller than the reference volume.
AI-generated editorial illustration.

On a standard Wechsler sheet, where 100 is the population mean and 15 is one standard deviation, a General Memory Quotient of 115 and a Processing Speed Index of 110 do not look like trouble. Both sit in the “high average” band, comfortably ahead of most people tested. Yet the 63-year-old former computer programmer who produced those scores had come back to follow-up complaining of persistent memory difficulties.

The rest of his test sheet explains why those two numbers puzzled his doctors. On the Wechsler Adult Intelligence Scale (WAIS-III), administered roughly five years after surgical clipping of a ruptured anterior communicating artery (ACoA) aneurysm, he scored 144 for Full-Scale IQ, 156 for Verbal IQ, and 150 for Working Memory. Against his Full-Scale IQ of 144, his 115 on the Wechsler Memory Scale–Revised (WMS-R) is a 29-point gap, and his 110 in processing speed is a 34-point gap. Clinical convention treats an intra-individual discrepancy above 15 points as significant.

The case was published online on 29 December 2025 in the Journal of Clinical Imaging Science by Shunji Mugikura of Tohoku University and Naoko Mori of Akita University. Ten months after the operation, high-resolution 3D MRI had already shown small, sharply placed infarctions that matched the split on his score sheet.

What the clip cut off, and what collateral blood saved

Surgery near an anterior communicating artery aneurysm runs beside tiny perforating vessels that rarely appear clearly on thick-slice acute scans. One is the subcallosal artery, usually unpaired, which rises from the ACoA in an S-shaped curve and supplies eight medial structures per side, including the columns of the fornix—core fibres of the Papez memory circuit—as well as the paraterminal gyri and the anterior cingulate gyrus. Next to it run the paired recurrent arteries of Heubner (RAH), which arise near the junction of the A1 and A2 segments and feed lateral basal forebrain and striatal regions.

On the patient’s 10-month 3D T1- and T2-weighted MRI, the subcallosal territory had suffered bilateral damage: both columns of the fornix had cavitated into small fluid-like hollows, and the paraterminal gyri and subcallosal areas were infarcted on both sides. Yet three distal targets of the same artery—the rostrum and genu of the corpus callosum, and the anterior cingulate gyrus—remained structurally intact, which the authors attribute to probable collateral flow from the A2 segment of the anterior cerebral artery.

A second, asymmetric injury appeared in the right RAH territory, involving the nucleus accumbens, putamen, globus pallidus, and the anteroinferior anterior limb of the internal capsule. Here too, one neighbour survived: the head of the caudate nucleus was spared on both sides, consistent with the complementary blood supply often provided by lateral striate arteries.

Two kinds of memory pulling apart

What makes the neuropsychological table striking is not only the 29-point gap between Full-Scale IQ and General Memory, but the divide inside memory itself. On the WAIS-III, his Working Memory Index reached 150 and his Verbal Comprehension 147; on the WMS-R, his Attention/Concentration index reached 131. Holding information actively in mind over seconds remained at the very superior level.

Episodic retention did not. His WMS-R Verbal Memory scored 112, Visual Memory 115, and Delayed Recall 112. In their 2025 report, Mugikura and Mori map those two selective drops onto two separate circuits—bilateral subcallosal infarction interrupting the Papez loop for episodic memory (alongside the basal forebrain cholinergic projections discussed in the group’s 2015 and 2020 studies), and right RAH infarction disrupting fronto-striatal pathways for processing speed.

Comparing this single patient with a 2020 SPECT series of ten ACoA-amnesia patients by the same group (selected for a discrepancy of at least 15 points between Full-Scale IQ and General Memory) clarifies why his scores stayed above 100. Those ten patients had Full-Scale IQs between 82 and 120 and General Memory scores between 58 and 92, with gaps of 16 to 43 points—and seven of the ten fell below the measurable floor of 50 on Delayed Recall. In that 2020 cohort, reduced blood flow in the right subcallosal gyrus, right rectal gyrus, and right anterior cingulate gyrus correlated with worse memory scores. Here, a 29-point gap sits right in the middle of that 16-to-43 range, yet spared anterior cingulate cortex and a much higher starting baseline kept the absolute numbers above the population mean.

What the case cannot isolate

Several limits keep this report from reading like a clean laboratory dissociation. First, no pre-operative IQ or memory test exists. Both the 29-point and 34-point gaps compare scores measured in the same testing session five years after surgery, using Verbal and Full-Scale IQ and the patient’s former career as a computer programmer as proxies for where his memory and speed once stood.

Second, roughly four years separate the 10-month 3D MRI from the five-year cognitive evaluation, and the scan itself could not inspect every structure equally. Metallic clip artifacts obscured the anterior commissure bilaterally, as well as the diagonal band of Broca and the bed nucleus of the stria terminalis on the right—structures marked as unreadable in the authors’ anatomical table.

Third, the MRI also recorded lesions outside the two named perforators: bilateral orbitofrontal infarcts (larger on the right) and an infarct in the right anterior temporal pole attributed to surgical manipulation through the right Sylvian fissure approach, on top of the initial haemorrhage from a ruptured aneurysm. (The paper also shows a minor internal slip, referring in text to six RAH territories while its anatomical table and figure legend list five.)

As a single retrospective case, the report cannot say how often postoperative memory loss goes undetected behind high baseline scores. What it shows concretely is a measurement blind spot: an automated flag set only at 85 or 100 passes over a patient who lost nearly two standard deviations relative to himself.

02 / A Note on Sources

A Note on Sources

I read the complete open-access text, tables, and figure captions of Shunji Mugikura and Naoko Mori, “Paradoxical giftedness and memory decline after anterior communicating artery aneurysm clipping: A high-resolution MRI case report,” Journal of Clinical Imaging Science (published online 29 December 2025, PMC12860287, DOI 10.25259/JCIS_179_2025). All WAIS-III and WMS-R scores (Table 2), anatomical ratings including the three clip-obscured structures and extra-territorial orbitofrontal and temporal-pole infarcts (Table 1), and the 10-month MRI and 5-year testing intervals come directly from that paper; the text’s mention of “six RAH territories” versus five listed columns in Table 1 and Figure 2 is also directly observed there. To cross-check the group’s dual-circuit framework and earlier patient series, I read in full Mugikura et al., Annals of Nuclear Medicine 34:220–227 (2020, PMC7033071, 10 patients, Table 1 and Spearman correlations in Table 2) and Mugikura & Takahashi, Neuroradiology 57:757–759 (2015, PMC4495264). For Mugikura et al., AJNR American Journal of Neuroradiology 35:2293–2301 (2014, PMC7965321 / PMID 25082820) and Mugikura et al., Japanese Journal of Radiology 36:81–91 (2018, PMID 29170982), only abstracts were available in the PubMed/PMC XML feeds, and those papers were not read in full.

03 / Sources

High-resolution 3D MRI and WAIS-III / WMS-R dissociation after ACoA clipping

Shunji Mugikura & Naoko Mori, Journal of Clinical Imaging Science (2025), alongside earlier anatomical and SPECT studies by the Tohoku/Akita group.

The case report

Earlier series and anatomy

How this was written: researched and drafted with the help of AI tools, starting from the sources listed here. Editorial responsibility for what is published is mine. Corrections: hello@liminalfinds.com.

If this was worth your time, you can support Liminal Finds (opens in a new tab).