Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2026 May 8;28(4):1175–1187. doi: 10.1002/epd2.70261

Predicting post‐surgical visuospatial memory performance in people with drug‐resistant epilepsy

Emma Colucci 1,2, Arnaud Saj 1, Simon Charbonneau 3, Carole Denault 3, Annie Malenfant 3, Isabelle Rouleau 2,4,5, Dang Khoa Nguyen 2,5,6, Olivier Boucher 1,2,3,✉
PMCID: PMC13499281  PMID: 42101155

Abstract

Objective

Although verbal memory decline has been consistently reported following surgery for temporal lobe epilepsy, especially when surgery is performed in the language‐dominant hemisphere and targets mesial temporal structures, the relation between the location and the hemisphere of surgery on visuospatial memory, as well as the factors predicting visuospatial memory decline, remains less clear. This study first examines the impact of epilepsy surgery location and laterality on visuospatial and verbal memory performance and then explores the predictors of memory change following surgery.

Methods

We retrospectively collected data from 75 individuals who completed the Brief Visuospatial Memory Test‐Revised and the Rey Auditory Verbal Learning Test as part of their pre‐ and post‐surgical neuropsychological assessments at our center.

Results

Surgery for mesial temporal lobe epilepsy in the dominant hemisphere was associated with poorer performance on all memory outcomes, whereas mesial temporal lobe epilepsy surgery in the non‐dominant hemisphere selectively impaired delayed recall of visuospatial material. A presurgical memory profile combining better visuospatial memory and poorer verbal memory, and the recurrence of seizures after surgery, best predicted visuospatial memory decline after the procedure, whereas verbal memory decline was predicted by better preoperative verbal memory performance in addition to surgery in the dominant hemisphere.

Significance

These findings emphasize the relevance of considering presurgical visuospatial and verbal memory performance when attempting to predict visuospatial memory decline following surgery. Limitations to our study include potential selection bias, variability in testing parameters, and a relatively small sample size.

Keywords: epilepsy surgery, hemispheric specialization, material‐specific memory impairment, memory specialization index, mesiotemporal lobe epilepsy (mTLE), non‐verbal memory

Abbreviations

ASMs

Antiseizure medications

BVMT‐R

Brief Visuospatial Memory Test‐Revised

EZ

Epileptogenic zone

mTLE

Mesial temporal lobe epilepsy

RAVLT

Rey Auditory Verbal Learning Test

TLE

Temporal lobe epilepsy

Key points.

  1. Individuals with mTLE showed greater visuospatial memory decline than those with non‐mTLE.

  2. Dominant mesiotemporal regions seem to support both verbal and visuospatial memory functions.

  3. Decline in verbal memory after surgery was predicted by preoperative verbal memory.

  4. Decline in visuospatial memory after surgery was predicted by the memory specialization index and post‐surgical seizure recurrence.

  5. The memory specialization index was a stronger predictor of visuospatial memory decline than preoperative performance alone.

1. INTRODUCTION

Temporal lobe epilepsy (TLE) is consistently associated with memory impairment, especially when the mesial structures are involved. 1 The laterality of the epileptogenic zone (EZ) has been shown to influence the type of memory deficits observed. Specifically, mesial temporal lobe epilepsy (mTLE) originating in the language‐dominant (generally left) hemisphere is closely associated with verbal memory impairments. 1 , 2 In contrast, mTLE in the non‐dominant hemisphere tends to be associated with non‐verbal memory deficits, although this result is not as robust. In fact, while some authors have reported visuospatial memory impairments in right TLE, 3 others have documented such impairments in left TLE as well. 4

The impact of epilepsy surgery on memory performance also varies according to the location and laterality of the intervention. For verbal memory, most studies report postoperative decline following left temporal lobe surgery. 5 , 6 Nonetheless, some patients who undergo surgery for right TLE also experience verbal memory decline. 6 , 7 Conversely, while less common, some individuals, particularly those for whom the procedure is in the non‐dominant hemisphere, show no postoperative change or even improvement in verbal memory performance. 5

The effects of epilepsy surgery are less clear regarding non‐verbal memory. While some studies report worse visuospatial memory after surgery for right mTLE, 4 others find no significant change 8 or even report improvements. 6 , 8 Visuospatial memory decline has also been observed after left‐sided resection. 9 A systematic review of the literature 3 suggests that the association between right mTLE and non‐verbal memory becomes stronger after surgery in comparison to before surgery.

Most studies examining the predictors of memory decline following epilepsy surgery have focused on verbal memory. Among the most consistent predictors of verbal memory deterioration across studies are better preoperative verbal memory performance, 10 , 11 , 12 , 13 left laterality of surgery, 11 , 12 , 13 , 14 and shorter duration of disease (related to an older age at onset). 11 , 15 Resections of mesiotemporal structures have also been linked to greater postoperative verbal memory decline. 12 , 13 Other factors have been associated with verbal memory decline, although less consistently: presence of hippocampal sclerosis, 16 , 17 post‐surgical seizure frequency, 18 , 19 sex (male 20 or female 21 ), age at time of surgery (younger 11 or older 10 , 15 ).

Much less is known about the predictors of visuospatial memory change after epilepsy surgery. In the few studies examining variables associated with visuospatial memory decline, side of surgery appears to be one of the most frequently investigated, but still no clear consensus has been reached on its impact. Indeed, some have reported an association between right hemisphere surgery and non‐verbal memory decline 22 , 23 whereas most others have found inconsistent or weak associations. 13 , 24 Higher preoperative non‐verbal memory scores have also been linked to greater postoperative visuospatial memory decline. 4 , 23 The evidence is mixed for the effects of age and duration of disease. 3 , 4 , 23 , 25 , 26 Higher postoperative seizure frequency appears to predict poorer non‐verbal memory outcomes, 22 , 25 , 27 although verbal and non‐verbal memory measures are frequently mixed in the analyses, making it unclear whether non‐verbal memory is directly affected.

There is increasing evidence from functional neuroimaging studies that preoperative functional asymmetry in the hippocampi is also predictive of postoperative memory decline. For example, Bonelli et al. 4 reported that stronger activation in the left than right anterior hippocampus during word encoding was associated with greater postoperative verbal memory decline, whereas stronger right than left anterior hippocampus activation during face encoding was associated with greater postoperative visual memory decline in individuals with right TLE. Powell and colleagues 28 also observed that those showing greater non‐dominant hippocampal activation during memory encoding tended to experience greater non‐verbal memory decline following non‐dominant surgery. Whether asymmetry in neuropsychological test results can also help predict postoperative memory decline remains unknown. Using data from a large sample of 692 patients with hippocampal sclerosis, Baxendale and Thompson 23 demonstrated that a memory specialization index — calculated by subtracting normalized visuospatial memory scores from verbal memory scores — was better at identifying patients with a right EZ than verbal or visuospatial memory performance alone. However, the ability of such a memory specialization index to predict postoperative memory performance is uncertain.

A systematic review of the literature 3 highlights that studies often lack comparisons between individuals with mTLE and those with other types of epilepsies. Thus, it is currently unclear whether variations in visuospatial memory performance are attributable solely to the laterality or also to the location of the EZ.

In summary, whereas the impact of epilepsy surgery on verbal memory has been extensively investigated, there remains uncertainty with regards to the postoperative non‐verbal memory outcomes in individuals undergoing epilepsy surgery. The present study first examines the impact of EZ location and laterality on visuospatial memory, as assessed with the Brief Visuospatial Memory Test – Revised (BVMT‐R), in patients undergoing neurosurgery for drug‐resistant focal epilepsy. A second aim is to explore the predictors of visuospatial memory change after epilepsy surgery. Verbal memory performance is also examined for comparison purposes.

2. MATERIALS AND METHODS

2.1. Participants

All neuropsychological files from patients who underwent a first epilepsy surgery at our center since the introduction of the BVMT‐R in our neuropsychological test battery were retrospectively screened (from February 2009 to May 2025). All patients were adults and aged <65 years old at time of surgery. Patients with high‐grade tumors and/or with intellectual disability were not considered. Out of the 126 individuals meeting these criteria, data from all patients (n = 78) who had performed the BVMT‐R as part of both pre‐ and post‐surgical assessments were compiled (22 patients were not assessed with the BVMT‐R at presurgical assessment mainly due to a gradual integration of the test during the first few years of its integration into our test battery, an additional 23 did not undergo post‐surgical assessment either because they were not referred by their treating neurologist or neurosurgeon, or because they were not available, and another 3 patients did not perform the BVMT‐R at their post‐surgical assessment). Three patients were excluded due to invalid performance confirmed by validity tests at the post‐surgical assessment. The patients retained in the final sample were comparable to those excluded for the aforementioned reasons for most sociodemographic and epilepsy‐related variables (Table S1), suggesting that our final sample is representative of patients operated at our center during the period covered by our study. Details about the pathology underlying epilepsy and surgical procedures of the final 75 participants are reported in Table S2. Language dominance was determined either by Wada or fMRI when atypical hemispheric representation was suspected.

2.2. Procedure

Most presurgical neuropsychological assessments were conducted at the bedside of patients in the video‐EEG monitoring unit, whereas a majority of post‐surgical assessments were conducted at our outpatient clinic, as per medical request. Patients were assessed on a comprehensive battery of neuropsychological tests including the memory tests that are the object of the present study. For each assessment, the examiner ensured beforehand that the patient's condition was not altered by seizures occurring in the preceding hours. Time of post‐surgical assessments varied according to the time of medical request and neuropsychologist's and patient's availability. In cases where multiple postoperative evaluations were conducted, the earliest assessment was retained for the present study. These were generally conducted ≥6 months after surgery according to the International League Against Epilepsy – Neuropsychology Task Force recommendation, 29 except for one patient who was assessed only 1.5 months after laser interstitial thermal therapy (range of postoperative delays = 1.5 to 114 months; 48% were seen within the first year after surgery).

This study was approved by the CHUM Research Center ethics committee (Project #20.110).

2.3. Neuropsychological tests

Visuospatial memory was assessed using the BVMT‐R. 30 This test was developed to address the practice effects on memory tests and to be analogous to verbal memory tests that use learning, recall, and recognition trials. The BVMT‐R is named as a potential tool for evaluating non‐verbal memory in people with epilepsy by the National Institute of Neurological Disorders and Stroke (NINDS). 31 During three learning trials, participants are shown a matrix of six geometric figures displayed in a 2 × 3 array for 10 s per trial. Afterwards, they must draw the correct figures in the correct locations on a blank page from memory. Each figure is given 1 point for accuracy and 1 point for correct location, for a maximal score of 12 points per trial. A Total Learning score is computed by summing the scores from the three trials. Twenty‐five minutes after the third learning trial, participants must once again draw the six figures in the correct locations from memory (Delayed Recall). Immediately afterwards, a Recognition trial is administered, during which the participants are shown 12 figures (6 targets, 6 distractors), one at a time, and must indicate whether each figure is a target or a distractor.

Verbal memory was assessed using the Rey Auditory Verbal Learning Test (RAVLT). 32 The RAVLT is among the tests recommended by the NINDS 31 to assess verbal memory. The examiner reads a 15‐word list (List A), and the participant recalls as many words as possible over five learning trials. The procedure is repeated with a 15‐word distractor list (List B) but over only one trial. Then, the participant recalls as many words as possible from List A (Immediate recall). Thirty minutes later, another recall trial is performed (Delayed recall), which is followed in our adapted procedure by a yes/no recognition trial with 50 words (15 from List A, 15 from List B, and 20 new distractors) where the participant indicates whether the word was in List A or not. The French or English version was used, according to the patient's native language or, if their native language was neither, the language in which they were more comfortable.

For the aims of this study, the sum of the learning trials (total learning), as well as delayed recall trials, was selected as a main measure of memory performance. Scores were converted to z‐scores adjusting for the effect of age using published age‐dependent normative data for the BVMT‐R, 33 as well as age and gender corrected normative data recently developed by our group for the RAVLT in a Quebec population (like our sample). 34 All z‐scores below −3.0 were considered outliers and were therefore recoded to −3.0. For both memory tasks, alternative versions of the tests were generally used for post‐surgical assessments, especially for shorter time intervals between the two assessments.

2.4. Statistical analysis

Participants were grouped based on laterality (dominant vs. non‐dominant hemisphere) and location (mTLE vs. non‐mTLE) of the surgery: dominant mTLE, dominant non‐mTLE, non‐dominant mTLE and non‐dominant non‐mTLE. Laterality of surgery was classified as dominant if it was conducted in the language‐dominant hemisphere and non‐dominant if it involved the other hemisphere. Location of surgery was considered as mesiotemporal (mTLE) if it involved mesial temporal lobe structures, even when other brain regions were also involved. Surgeries that did not involve mesiotemporal structures were classified as non‐mTLE. Chi‐square (categorical variables) and ANOVA (continuous variables) tests were conducted to compare descriptive variables between the groups. When the omnibus test was significant, post‐hoc pairwise comparisons were conducted using Bonferroni correction.

To examine whether location (mesiotemporal vs. non‐mesiotemporal) and laterality (dominant vs. non‐dominant hemisphere) predicted changes in memory performance from pre‐ to post‐surgery, a 2 × 2 × 2 mixed general linear model (GLM) was conducted, with time (pre, post) as the within‐subject factor and EZ location and laterality as between‐subject factors for the four main outcomes (Total Learning and Delayed Recall on the BVMT‐R and RAVLT). When significant effects were found, post‐hoc pairwise comparisons of estimated marginal means were conducted and adjusted for multiple comparisons using the Bonferroni correction.

Afterward, predictors of postoperative change in memory were examined. Pearson correlations were conducted to investigate the relationship between predictors and change in performance (post‐surgery vs. presurgery) for the four main memory outcomes. Because Engel classes were unevenly distributed across the groups (see Table 1), we simplified the classification into a binary predictor (seizure free [Class I] vs. not seizure free [Classes II, III, and IV]) to improve statistical power. The following variables were investigated: sex (male: 1; female: 2), education (yrs), age at diagnosis (yrs), duration of epilepsy (yrs), age at surgery (yrs), change in the number of antiseizure medications (ASMs) from pre‐ to post‐surgery (Δ #of ASMs), seizure control following surgery (Engel classification I vs. ≥ II at time of post‐surgical assessment), EZ laterality (dominant vs. non‐dominant), EZ location (mTLE vs. non‐mTLE), preoperative verbal and visuospatial memory score (sum of learning trials for analyses examining learning as the outcome; delayed recall trials for analyses examining delayed recall as the outcome), preoperative memory specialization index (preoperative RAVLT z‐score minus BVMT‐R z‐score; negative scores indicating advantage for visuospatial material whereas positive scores indicating advantage for verbal material). Predictors which were most often correlated with memory scores were included in stepwise multiple linear regression models.

TABLE 1.

Sample description.

Variable Dominant mTLE (n = 22) Dominant non‐mTLE (n = 13) Non‐dominant mTLE (n = 25) Non‐dominant non‐mTLE (n = 15) p
n (%) Mean (SD) n (%) Mean (SD) n (%) Mean (SD) n (%) Mean (SD)
Sex 0.366
Male 5 (22.7%) 6 (46.2%) 8 (32.0%) 7 (46.7%)
Female 17 (77.3%) 7 (53.8%) 17 (68.0%) 8 (53.3%)
Education (years) 12.4 (2.7) 12.5 (3.7) 12.1 (3.3) 13.2 (2.1) 0.730
Age at diagnosis 21.4 (17.5) 14.7 (10.8) 15.0 (11.1) 17.1 (10.1) 0.339
Age at surgery 39.8 (13.6) 32.4 (11.4) 34.2 (9.8) 34.0 (7.9) 0.174
Duration of epilepsy (years) 18.5 (15.1) 17.7 (14.8) 19.2 (14.5) 16.9 (10.4) 0.965
Delay between presurgical assessment vs. post‐surgical assessment (months)

34.8 (27.3);

13.4–118.8 (24.0)

21.1 (9.5);

8.7–41.6 (19.0)

32.6 (28.9);

8.8–125.2 (24.7)

28.7 (15.1);

12.4–70.2 (26.2)

 0.390
Delay between surgery and post‐surgical assessment (months)

18.7 (23.3);

5.6–114.0 (10.6)

12.0 (8.0);

5.5–36.1 (10.8)

20.4 (21.5);

1.5–114.5 (16.1)

14.7 (9.4);

5.5–41.6 (11.5)

 0.543
Number of ASMs before surgery 2.1 (0.8) 2.4 (0.9) 1.8 (0.8) 2.0 (1.2) 0.320
Number of ASMs after surgery 2.0 (0.8) 2.7 (0.9) 1.7 (0.7) 2.0 (0.7) 0.002
Engel classification 0.436
1 12 (54.5%) 9 (69.2%) 15 (60.0%) 12 (80.0%)
2 7 (31.8%) 1 (7.7%) 6 (24.0%) 0 (0.0%)
3 2 (9.1%) 2 (15.4%) 3 (12.0%) 3 (20.0%)
4 1 (4.5%) 1 (7.7%) 1 (4.0%) 0 (0.0%)
Open in a new tab

Note: Groups are defined according to seizure focus laterality (dominant vs. non‐dominant hemisphere) and epilepsy type (mTLE and non‐mTLE). “Presurgical assessment” refers to the neuropsychological evaluation conducted prior to surgery, whereas “post‐surgical assessment” refers to the first follow‐up evaluation conducted after surgery. Range (median) are additionally provided for post‐surgical timing variables to reflect variability in follow‐up intervals. p‐value of chi‐square (sex and Engel classification) and ANOVA (other variables) tests are reported.

Abbreviations: ASM = antiseizure medication; mTLE = mesiotemporal lobe epilepsy; SD = standard deviation.

Assumptions were verified for all conducted tests. Considering the small sample size, the minimum number of participants per group needed for multiple linear regression was not obtained, which is why these analyses were considered exploratory.

All statistical analyses were performed using IBM SPSS Statistics software (version 28). The significance level was set at p < 0.05.

3. RESULTS

Descriptive statistics according to the location and laterality of epilepsy surgery are reported in Table 1. Groups only differed on the number of ASMs taken after surgery: post‐hoc comparisons using Tukey's HSD test revealed that the dominant non‐mTLE group was prescribed more ASMs after surgery than the dominant mTLE (p = .049) and the non‐dominant mTLE groups (p < .001).

3.1. Effect of surgery on performance according to location and laterality of surgery

The results of the GLMs testing for the effects of surgery location and laterality on memory performance are presented in Table 2.

TABLE 2.

General linear model results for memory outcomes (BVMT‐R and RAVLT).

Effect Group Presurgery mean (SD) Post‐surgery mean (SD) F (df1, df2) p η2p
BVMT‐R Total Learning Time All −0.54 (0.16) −0.55 (0.16) 0.01 (1, 71) .920 < .001
Time × Laterality 0.26 (1, 71) .609 .004
Dominant −0.33 (0.24) −0.42 (0.23)
Non‐dominant −0.75 (0.22) −0.69 (0.21)
Time × Location 4.32 (1, 71) .041* .057
mTLE −0.54 (0.20) −0.85 (0.20) .057†
Non‐mTLE −0.54 (0.26) −0.25 (0.25)
Time × Laterality × Location 4.48 (1, 71) .038* .059
Dominant mTLE −0.19 (0.20) −0.87 (0.28) 0.009**
Non‐dominant mTLE −0.90 (0.27) −0.84 (0.26)
Dominant non‐mTLE −0.47 (0.38) 0.04 (0.36)
Non‐dominant non‐mTLE −0.60 (0.35) −0.55 (0.34)
BVMT‐R Delayed Recall Time All −0.54 (0.17) −0.68 (0.18) 0.84 (1, 68) .364 .012
Time × Laterality 0.08 (1, 68) .780 .012
Dominant −0.54 (0.26) −0.63 (0.27)
Non‐dominant −0.55 (0.23) −0.72 (0.25)
Time × Location 2.87 (1, 68) .095† .040
mTLE −0.58 (0.21) −0.97 (0.23)
Non‐mTLE −0.50 (0.27) −0.38 (0.29)
Time × Laterality × Location 0.25 (1, 68) .620 .004
Dominant mTLE −0.48 (0.31) −0.91 (0.33)
Non‐dominant mTLE −0.68 (0.29) −1.04 (0.31)
Dominant non‐mTLE −0.59 (0.41) −0.36 (0.43)
Non‐dominant non‐mTLE −0.41 (0.37) −0.41 (0.39)
RAVLT Total Learning Time All −0.49 (0.15) −0.57 (0.15) 0.38 (1, 71) .540 .005
Time × Laterality 5.07 (1, 71) .028* .067
Dominant −0.46 (0.21) −0.81 (0.20) .054†
Non‐dominant −0.53 (0.20) −0.33 (0.20)
Time × Location 0.02 (1, 71) .889 < .001
mTLE −0.66 (0.18) −0.72 (0.18)
Non‐mTLE −0.32 (0.23) −0.42 (0.23)
Time × Laterality × Location 3.56 (1,71) .063† .048
Dominant mTLE −0.59 (0.26) −1.15 (0.26) 0.012*
Non‐dominant mTLE −0.74 (0.24) −0.29 (0.24) 0.032*
Dominant non‐mTLE −0.32 (0.34) −0.59 (0.26)
Non‐dominant non‐mTLE −0.32 (0.32) −0.37 (0.32)
RAVLT Delayed Recall Time All −0.62 (0.15) −0.84 (0.16) 4.15 (1, 68) .046* .058
Time × Laterality 2.86 (1, 68) .095† .040
Dominant −0.68 (0.22) −1.09 (0.23) 0.014*
Non‐dominant −0.55 (0.20) −0.59 (0.21)
Time × Location 2.35 (1, 68) .130 .033
mTLE −0.75 (0.18) −1.15 (0.19) 0.005**
Non‐mTLE −0.48 (0.23) −0.53 (0.25)
Time × Laterality × Location 14.83 (1,68) < .001** .179
Dominant mTLE −0.67 (0.26) −1.69 (0.28) < .001**
Non‐dominant mTLE −0.84 (0.25) −0.62 (0.26)
Dominant non‐mTLE −0.68 (0.35) −0.50 (0.37)
Non‐dominant non‐mTLE −0.84 (0.25) −0.62 (0.26)
Open in a new tab

Note: Time = pre‐ vs. post‐surgery; Laterality = dominant vs. non‐dominant hemisphere; Location = mTLE vs. non‐mTLE. The “Time” effect examines whether there is a significant difference between pre‐ and postoperative scores across the whole sample. “Time × Laterality” compares changes between dominant and non‐dominant hemispheres, regardless of the lobe involved. “Time × Location” compares changes between mTLE and non‐mTLE participants, regardless of hemisphere. Finally, “Time × Laterality × Location” evaluates the combined effect of both hemisphere and mesiotemporal involvement. **p < 0.01; *p < 0.05; †p < 0.10.

Abbreviations: BVMT‐R, Brief Visuospatial Memory Test – Revised; RAVLT, Rey Auditory Verbal Learning Test; ×, average Z score;

SD, standard deviation; η2p, Partial Eta Squared.

For BVMT‐R Total Learning (see Figure 1A), significant effects were found for Time × Location: mTLE patients showed generally lower scores than non‐mTLE patients and a trend toward lower scores after surgery (p = 0.057). The Time × Laterality × Location interaction was also significant: the dominant mTLE group showed significantly lower scores after surgery (M = −0.87 ± 0.28) than before surgery (M = −0.19 ± 0.29; p = 0.009), whereas performance remained stable in time in the other groups. No significant main effects or interactions were observed for BVMT‐R Delayed Recall (see Figure 1B), but the Time × Location interaction fell short of statistical significance (p = 0.095). Follow‐up pairwise comparisons indicated that patients with mTLE had significantly worse performance after surgery (M = −0.97 ± 0.23) than before surgery (M = −0.58 ± 0.21; p = 0.036); this was not observed in patients with non‐mTLE.

FIGURE 1.

FIGURE 1

Open in a new tab

Visuospatial and verbal memory performance before and after surgery for Dominant mTLE (blue), Non‐dominant mTLE (red), Dominant non‐mTLE (green), and Non‐dominant non‐mTLE (yellow). **p < 0.01; *p < 0.05.

For RAVLT Total Learning (see Figure 1C), the Time × Laterality interaction was statistically significant, with patients operated in the dominant hemisphere showing a trend for worse performance postoperatively (M = −0.81 ± 0.21) than preoperatively (M = −0.46 ± 0.21; p = 0.054). Although the Time × Laterality × Location interaction fell short of statistical significance, follow‐up simple effect analyses showed significant decline from pre‐ to post‐surgery for dominant‐mTLE (from M = −0.59 to −1.15, p = 0.012), whereas patients in the non‐dominant mTLE group saw an increase in performance (from M = −0.74 to −0.29, p = 0.032) and non‐mTLE groups remained stable over time. Moreover, dominant mTLE surgery (M = −1.15 ± 0.26) was associated with worse scores post‐surgery than non‐dominant mTLE (M = −0.46 ± 0.34; p = 0.019). For RAVLT Delayed Recall (see Figure 1D), the Time × Laterality × Location interaction was significant, and post‐hoc pairwise comparisons indicated that mTLE surgery in the dominant hemisphere (M = −1.69 ± 0.28) was linked to significantly worse postoperative scores than surgery for mTLE in the non‐dominant hemisphere (M = −0.62 ± 0.26; p = 0.007), and for non‐mTLE in the dominant hemisphere (M = −0.50 ± 0.37; p = .012). To add, performance of the dominant mTLE group was significantly worse after surgery (M = −1.69 ± 0.28) than before surgery (M = −0.67 ± 0.26; p < .001), but not in the other groups.

3.2. Predictors of postoperative change in memory performance

Pearson correlations between potential predictors and change in memory performance following surgery are reported in Table S3. Correlations between predictors showing significant associations with memory performance, along with results from the linear regression analyses, are reported in Table 3. The predictors included in the regression models were seizure recurrence after surgery, change in number of ASMs, EZ laterality, EZ location, preoperative BVMT‐R score, preoperative RAVLT score, and memory specialization index.

TABLE 3.

Association between predictors and change in performance after epilepsy surgery for all participants (N = 75).

BVMT‐R learning BVMT‐R delayed recall RAVLT learning RAVLT delayed recall
r Regression model r Regression model r Regression model r Regression model
F‐value adj. R 2 Std. B F‐value adj. R 2 Std. B F‐value adj. R 2 Std. B F‐value adj. R 2 Std. B
21.4** 0.46 18.3** 0.34 14.2** 0.27 6.74** 0.14
Seizure recurrence −0.37** −0.36** −0.34** −0.29** −0.01 −0.06
Δ ASMs −0.06 −0.15 −0.22† −0.16
Pre‐op BVMT‐R −0.49** −0.35** −0.21† −0.20†
Pre‐op RAVLT 0.08 −0.30** 0.16 −0.41** −0.43** −0.27* −0.29*
Memory specialization index 0.57** 0.64** 0.53** 0.49** −0.16 −0.04
EZ location −0.22† −0.20† 0.03 −0.13
EZ laterality −0.12 0.02 −0.31** −0.32** −0.29* −0.30**
Open in a new tab

Note: Change in performance is calculated by subtracting preoperative z‐score to postoperative z‐score. Seizure recurrence indicates an Engel class ≧ 2 at time of assessment. Change in number of antiseizure medications was calculated as the amount of ASMs taken after surgery – the amount of ASMs taken before surgery (a positive score indicated an increase in ASMs, a negative score indicated a decrease in ASMs). Memory specialization index was computed by subtraction the BVMT‐R score from the RAVLT score, a positive score indicated verbal dominance, a negative score indicated non‐verbal dominance. EZ location refers to grouping patients according to mesiotemporal vs. extra‐mesiotemporal lobe epilepsy; EZ laterality refers to grouping patients according to surgery in the dominant vs. non‐dominant hemisphere. **p < 0.01; *p < 0.05; †p < 0.10.

Abbreviations: adj. R2, adjusted R squared; BVMT‐R, Brief Visuospatial Memory Test – Revised; r, Pearson coefficient; RAVLT, Rey Auditory Verbal Learning Test; Std. B, standard beta coefficient; ΔASM, change in the number of antiseizure medications.

All overall multiple linear regression models predicting change in memory performance after surgery among patients were statistically significant. For BVMT‐R Total Learning, having seizures after surgery, better preoperative RAVLT score, and having a memory specialization index indicating dominance for visuospatial compared to verbal memory remained significant predictors of performance decline. For BVMT‐R Delayed Recall, seizures after surgery and having a memory specialization index showing dominance for visuospatial memory remained the only significant predictors of postoperative decline in performance in regression models. For RAVLT Total Learning and Delayed Recall, having better preoperative verbal memory performance and seizures originating in the dominant hemisphere were the only predictors of postoperative memory decline. Given the clinical relevance of memory outcomes in mTLE specifically, we further explored regression models within three groups: all individuals with mTLE, those with dominant‐mTLE, and those with non‐dominant mTLE (Table S4). As shown in Table S4, similar predictors of memory outcomes were identified across these three subgroups. This consistency suggests a degree of robustness in the regression models.

4. DISCUSSION

The first aim of this study was to examine the impact of the location and laterality of the EZ and of epilepsy surgery on visuospatial memory, using the BVMT‐R, in patients with drug‐resistant focal epilepsy. We found that, on average, individuals with mTLE experienced greater postoperative visuospatial memory decline than those with non‐mTLE. In fact, those who underwent surgery for dominant mTLE showed a decline in visuospatial learning, whereas delayed recall of visuospatial information was impacted by mTLE surgery regardless of the side of surgery. Visuospatial memory remained stable in groups of patients with non‐mTLE. In comparison, verbal memory learning and delayed recall were selectively affected by mTLE surgery in the dominant hemisphere, whereas the other groups showed no change or even improved after surgery, consistent with previous results. 5

Our results highlight the crucial role of mTLE structures in both verbal and non‐verbal memory performance 35 , 36 , 37 but challenge the traditional view that non‐verbal memory impairment is primarily associated with right or non‐dominant TLE. Indeed, the effects of dominant mTLE and its surgery on both verbal and visuospatial memory suggest that dominant mesiotemporal structures engage a broader network that supports both verbal and non‐verbal memory functions. This is consistent with previous work that highlighted bilateral neural networks involved in non‐verbal memory, as well as stronger functional reorganization mechanisms compared to verbal memory, 38 , 39 and aligns with the view that non‐verbal memory does not mirror verbal memory. 40

Even though significant effects of EZ location and laterality were found at the group level, memory profiles tended to be heterogeneous within groups. Previous studies had already reported that many individuals with TLE have “atypical” memory profiles (e.g., normal memory, bilateral deficits, contralateral deficits, memory deficits related to low IQ 37 ). In our sample, at least 15% of participants in each group had preoperative deficits in both verbal and non‐verbal memory. This heterogeneity reflects that, in some cases, the effects of TLE on memory are not perfectly lateralized. Although postoperative decline differed more consistently according to the location and laterality of the EZ, heterogeneity was still observed. In individuals with dominant mTLE, 19.2% declined in both verbal and non‐verbal learning, but only 1 participant (3.8%) declined in both verbal and non‐verbal delayed recall. In contrast, participants in the other three groups often showed a decline in only one type of memory, or no decline at all. Our data also revealed that, following surgery targeting the dominant mesiotemporal structures, more individuals experienced decline in both verbal and non‐verbal learning but not in delayed recall. One possible explanation relates to the critical role of the hippocampus in memory consolidation and retrieval. 25 , 40 We hypothesize that resection of the lesioned hippocampus allows for compensatory recruitment of the contralateral hippocampus, which could account for the preservation of delayed recall. Heterogeneity of EZ location (e.g., lateral, insular, frontal, etc.) in our non‐mTLE groups or individual differences in memory networks and strategies could also account for the variability in memory performances.

The second objective of our study was to explore the predictors of postoperative visuospatial memory decline in people undergoing neurosurgery for drug‐resistant focal epilepsy. Our results suggest that postoperative visuospatial memory decline is influenced by presurgical memory performance – and especially by the difference between preoperative visuospatial and verbal performance (i.e., better visuospatial memory and poorer verbal memory associated with greater decline) – and by seizure recurrence after surgery. Comparatively, decline in verbal memory performance was more likely among individuals who had surgery in the dominant hemisphere and who had better preoperative verbal memory performance.

To our knowledge, this is the first study to explore the predictive value of a memory specialization index in a postoperative setting in people with focal epilepsy. Our findings point to a potentially noteworthy and novel finding: memory dominance for visuospatial over verbal material before surgery is a better predictor for postoperative visuospatial memory outcomes than presurgical visuospatial memory performance alone, whereas verbal memory decline is best predicted by presurgical verbal memory performance alone (i.e., independently of visuospatial memory). These observations should be considered cautiously due to the relatively small sample size, but they underscore the value of further investigation in larger samples. Determining memory dominance can be achieved by subtracting visuospatial memory score from verbal memory score (in z‐scores), as proposed by Baxendale and Thompson, 23 who showed heightened value of this memory specialization index to lateralize the EZ in people with TLE. This approach has mostly been used in a preoperative setting (e.g., 41 , 42 ). To our knowledge, this is the first study to explore the relationship between memory dominance and postoperative change. Our results suggest that individuals with a higher memory specialization index before surgery, indicating stronger advantage for visuospatial over verbal memory, are at greater risk of postoperative visuospatial decline following epilepsy surgery, regardless of whether the dominant or the non‐dominant mesiotemporal structures are targeted. A likely explanation is that patients with preserved verbal memory are more likely to compensate for visuospatial impairments using verbalization strategies, whereas those with weaker preoperative verbal memory cannot rely such strategies and depend solely on “pure” visual memory abilities that are affected by the surgical intervention. Conversely, verbal memory decline is better predicted by preoperative memory function than by the memory specialization index, suggesting that better visuospatial memory does not help compensate for verbal memory decline. Our findings highlight the importance of considering both verbal and non‐verbal memory when evaluating the risks of TLE surgery on visuospatial memory.

Poorer postoperative visuospatial memory outcome in patients who continue to experience seizures after surgery in comparison to patients who are seizure free after surgery is also consistent with previous studies. 22 , 27

Among the limitations of the present study is its retrospective design, which may have contributed to heterogeneity regarding the use of alternate forms and delays before post‐surgical assessment. Not all patients who have undergone epilepsy surgery at our center during the study period were included potentially because, although the neurologist and neurosurgeon usually refer patients for follow‐up evaluations, those experiencing difficulties were more likely to return for postoperative neuropsychological assessment, which may have led to selection bias and overrepresentation of patients with subjective cognitive complaints. Moreover, some patients may not have undergone surgery partly based on their presurgical neuropsychological assessment results, also leading to a possible selection bias. Furthermore, the variability in elapsed time between surgery and post‐surgical evaluations (as evidenced by large standard deviations – 8 to 23.3 months) could have influenced cognitive recovery or compensation through neuroplasticity. Similarly, although memory outcomes were not significantly associated with changes in ASM number, we cannot exclude the possibility that changes in ASM type constitutes a better indicator of cognitive effects. The relatively small sample size could also limit statistical power and generalizability of our findings. In addition, participants older than 65 years were excluded to reduce the expected age‐related cognitive decline, which further limits generalizability, despite the prevalence of drug‐resistant epilepsy in older adults being estimated to increase with an aging population. 43 Finally, the visuospatial memory test used has its own limitations which could account for a part of our results. Indeed, although the BVMT‐R is frequently used in clinical settings, 33 it has shown limited sensitivity to right mTLE before surgery in previous studies. 44 , 45 Similar concerns have previously been brought to attention regarding the validity of other non‐verbal memory tests involving stimuli (e.g., shapes, figures, faces, etc.) which can often be verbalized at different levels, thus engaging the dominant hemisphere. 46 Despite these limitations, we nevertheless found that different factors influence the impact of epilepsy surgery on RAVLT and BVMT‐R performance, with only the latter being affected by non‐dominant mTLE surgery. Further studies are needed to verify whether our results can be replicated using other tests involving less verbalizable material, such as locations and landscapes. 47 , 48

5. CONCLUSION

The present study suggests that visuospatial memory may be sensitive to the effects of mTLE surgery, regardless of the laterality of the EZ, whereas verbal memory appears to be more specifically affected by mTLE surgery in the dominant hemisphere. Postoperative visuospatial memory changes were more closely related to the preoperative difference between visuospatial and verbal memory, whereas preoperative verbal memory alone was the strongest predictor of post‐surgical verbal memory changes. Our results highlight the relevance of considering both visuospatial and verbal memory prior to surgery when estimating the impact of epilepsy surgery on visuospatial memory. A relative material‐specific lateralization rather than a hemispheric lateralization may better account for memory outcomes. The memory specialization index shows promising clinical utility, as its integration into presurgical assessments could improve prediction of post‐surgical memory decline and contribute to optimizing EZ localization. Preoperative memory profiles (verbal vs. visuospatial) may also help guide tailored cognitive rehabilitation. Although replication with larger samples and other non‐verbal memory measures is needed, our results provide a novel and promising contribution to clinical practice, particularly for presurgical evaluation and postoperative cognitive care.

AUTHOR CONTRIBUTIONS

Emma Colucci: formal analysis, writing – original draft, writing – review and editing. Arnaud Saj: supervision, writing – review and editing. Simon Charbonneau: conceptualization, data curation, methodology, writing – review and editing. Carole Denault: conceptualization, data curation, methodology, writing – review and editing. Annie Malenfant: conceptualization, data curation, methodology, writing – review and editing. Isabelle Rouleau: conceptualization, methodology, writing – review and editing. Dang Khoa Nguyen: conceptualization, funding acquisition, data curation, methodology, writing – review and editing. Olivier Boucher: conceptualization, funding acquisition, data curation, formal analysis, methodology, supervision, writing – review and editing.

CONFLICTS OF INTEREST STATEMENT

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Emma Colucci reports financial support was provided by Fonds de Recherche du Québec–Santé (FRQS). Olivier Boucher reports financial support was provided by Fonds de Recherche du Québec–Santé (FRQS). Dang Khoa Nguyen reports financial support was provided by Canadian Institutes of Health Research. The other authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Test Yourself.

  1. What was the most novel finding of the study regarding predictors of postoperative visuospatial memory decline?
    1. Preoperative visuospatial performance alone predicted decline.
    2. Duration of epilepsy was the strongest predictor.
    3. Memory dominance (visuospatial > verbal performance before surgery) better predicted decline than visuospatial performance alone.
    4. Age at surgery predicted visuospatial decline.
  2. What do the study results suggest about the role of dominant mesiotemporal structures?
    1. They support only verbal memory.
    2. They contribute to both verbal and visuospatial memory functions.
    3. They are not involved in memory processing.
    4. They mainly support short‐term memory only
  3. Which factors were associated with poorer postoperative visuospatial memory outcome?
    1. Memory Specialization Index.
    2. Presurgical visuospatial memory performance
    3. Seizure recurrence after surgery.
    4. Change in the number of antiseizure medications taken

Answers may be found in the supporting information .

Supporting information

Table S1. Comparisons between the patients included in the final study sample and those excluded due to missing presurgical or post‐surgical BVMT‐R data or invalid performance.

Table S2. Distribution of pathology types and surgical interventions according to location and laterality of surgery.

Table S3. Pearson correlations between potential predictors and change in performance from pre‐ to postoperative assessment for the four main memory outcomes for the overall sample (N = 75).

Table S4. Association between predictors and change in performance after epilepsy surgery for mesiotemporal epilepsy.

EPD2-28-1175-s002.docx (35.8KB, docx)

Appendix S1.

EPD2-28-1175-s001.docx (11.3KB, docx)

ACKNOWLEDGMENTS

We thank all participants for their time and contribution to this research.

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available from the corresponding author upon reasonable request.

REFERENCES

  • 1. Alessio A, Damasceno BP, Camargo CHP, Kobayashi E, Guerreiro CAM, Cendes F. Differences in memory performance and other clinical characteristics in patients with mesial temporal lobe epilepsy with and without hippocampal atrophy. Epilepsy & Behavior. 2004;5(4):22–27. 10.1016/j.yebeh.2003.10.010 [DOI] [PubMed] [Google Scholar]
  • 2. Phuong TH, Houot M, Méré M, Denos M, Samson S, Dupont S. Cognitive impairment in temporal lobe epilepsy: Contributions of lesion, localization, and lateralization. Journal of Neurology. 2020;268:1443–1452. 10.1007/s00415-020-10307-6 [DOI] [PubMed] [Google Scholar]
  • 3. Bentvelzen AC, Kessels RPC, Badcock NA, Savage G. The impact of right temporal lobe epilepsy on nonverbal memory: meta‐regression of stimulus‐ and task‐related moderators. Neuropsychology Review. 2022;32:537–557. 10.1007/s11065-021-09514-3 [DOI] [PubMed] [Google Scholar]
  • 4. Bonelli SB, Powell RHW, Yogarajah M, Samson RS, Symms MR, Thompson PJ, et al. Imaging memory in temporal lobe epilepsy: predicting the effects of temporal lobe resection. Brain. 2010;133(4):1186–1199. 10.1093/brain/awq006 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5. Bremm FJ, Hendricks MPH, Bien CG, Grewe P. Pre‐ and postoperative verbal memory and executive functioning in frontal versus temporal lobe epilepsy. Epilepsy & Behavior. 2019;101:106538. 10.1016/j.yebeh.2019.106538 [DOI] [PubMed] [Google Scholar]
  • 6. Sherman EM, Wiebe S, Fay‐McClymont TB, Tellez‐Zenteno J, Metcalfe A, Hernandez‐Ronquillo L, et al. Neuropsychological outcomes after epilepsy surgery: systematic review and pooled estimates. Epilepsia. 2011;52(5):857–869. 10.1111/j.1528-1167.2011.03022.x [DOI] [PubMed] [Google Scholar]
  • 7. Jutila L, Äikiä M, Immonen A, Mervaala E, Alafuzoff I, Kälviäinen R. Long‐term memory performance after surgical treatment of unilateral temporal lobe epilepsy (TLE). Epilepsy Research. 2014;108(7):1228–1237. 10.1016/j.eplepsyres.2014.05.002 [DOI] [PubMed] [Google Scholar]
  • 8. Dulay MF, Levin HS, York MK, Mizrahi EM, Verma A, Goldsmith I, et al. Predictors of individual visual memory decline after unilateral anterior temporal lobe resection. Neurology. 2009;72(21):1837–1842. 10.1212/WNL.0b013e3181a71132 [DOI] [PubMed] [Google Scholar]
  • 9. Sone D, Ahmad M, Thompson PJ, Baxendale S, Vos SB, Xiao F, et al. Optimal surgical extent for memory and seizure outcome in temporal lobe epilepsy. Annals of Neurology. 2022;91(1):131–144. 10.1002/ana.26266 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10. Baxendale S, Thompson P, Harkness W, Duncan J. Predicting memory decline following epilepsy surgery: a multivariate approach. Epilepsia. 2006;47(11):1887–1894. 10.1111/j.1528-1167.2006.00810.x [DOI] [PubMed] [Google Scholar]
  • 11. Gross WL, Swanson SJ, Helfand AI, Pillay SB, Humphries CJ, Raghavan M, et al. Predicting memory decline from left temporal lobe epilepsy surgery using preoperative fMRI: a multicenter study. NeuroImage: Clinical. 2025;46:103804. 10.1016/j.nicl.2025.103804 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12. Ljunggren S, Andersson‐Roswall L, Imberg H, Samuelsson H, Malmgren K. Predicting verbal memory decline following temporal lobe resection for epilepsy. Acta Neurologica Scandinavica. 2019;140(5):312–319. 10.1111/ane.13146 [DOI] [PubMed] [Google Scholar]
  • 13. Parra‐Díaz P, García‐Casares N. Memory assessment in patients with temporal lobe epilepsy to predict memory impairment after surgery: a systematic review. Neurología. 2019;34(9):596–606. 10.1016/j.nrleng.2017.02.011 [DOI] [PubMed] [Google Scholar]
  • 14. Wisniewski I, Wendling AS, Steinhoff BJ. Impact of side of lesion, seizure outcome and interictal epileptiform discharges on attention and memory after surgery in temporal lobe epilepsy. Epileptic Disorders. 2011;13(1):27–35. 10.1684/epd.2011.0417 [DOI] [PubMed] [Google Scholar]
  • 15. Cano‐López I, Vázquez JF, Campos A, Gutiérrez A, Garcés M, Gómez‐Ibáñez A, et al. Age at surgery as a predictor of cognitive improvements in patients with drug‐resistant temporal epilepsy. Epilepsy Research. 2017;70:10–17. 10.1016/j.yebeh.2017.03.002 [DOI] [PubMed] [Google Scholar]
  • 16. Li Y, Liu P, Lin Q, Zhou D, An D. Postoperative seizure and memory outcome of temporal lobe epilepsy with hippocampal sclerosis: a systematic review. Epilepsia. 2023;64(11):2845–2860. 10.1111/epi.17757 [DOI] [PubMed] [Google Scholar]
  • 17. Martin RC, Kretzmer T, Palmer C, Sawrie S, Knowlton R, Faught E, et al. Risk to verbal memory following anterior temporal lobectomy in patients with severe left‐sided hippocampal sclerosis. Archives of Neurology. 2002;59(12):1895–1901. 10.1001/archneur.59.12.1895 [DOI] [PubMed] [Google Scholar]
  • 18. Baxendale S, Thompson PJ, Sander JW. Neuropsychological outcomes in epilepsy surgery patients with unilateral hippocampal sclerosis and good preoperative memory function. Epilepsia. 2013;54(9):e131–e144. 10.1111/epi.12319 [DOI] [PubMed] [Google Scholar]
  • 19. Alpherts WC, Vermeulen J, van Rijen PC, da Silva FH, van Veelen CW. Verbal memory decline after temporal epilepsy surgery?: a 6‐year multiple assessments follow‐up study. Neurology. 2006;67(4):626–631. 10.1212/01.wnl.0000230139.45304.eb [DOI] [PubMed] [Google Scholar]
  • 20. Trenerry MR, Jack CR, Cascino GD, Sharbrough F, Ivnik RJ. Gender differences in post–temporal lobectomy verbal memory and relationships between MRI hippocampal volumes and preoperative verbal memory. Epilepsy Research. 1995;20(1):69–76. 10.1016/0920-1211(94)00060-A [DOI] [PubMed] [Google Scholar]
  • 21. Baxendale S. Sex differences in the pre and postoperative neuropsychological function of epilepsy surgery candidates. The Clinical Neuropsychologist. 2023;38(6):1441–1453. 10.1080/13854046.2023.2281706 [DOI] [PubMed] [Google Scholar]
  • 22. Baxendale S, Thompson PJ, Duncan JS. Neuropsychological function in patients who have had epilepsy surgery: a long‐term follow‐up. Epilepsy & Behavior. 2012;23(1):24–29. 10.1016/j.yebeh.2011.10.021 [DOI] [PubMed] [Google Scholar]
  • 23. Baxendale S, Thompson P. The association of cognitive phenotypes with postoperative outcomes after epilepsy surgery in patients with temporal lobe epilepsy. Epilepsy & Behavior. 2020;112:107386. 10.1016/j.yebeh.2020.107386 [DOI] [PubMed] [Google Scholar]
  • 24. Berger J, Oltmanns F, Holtkamp M, Benger T. Sex differences in verbal and nonverbal learning before and after temporal lobe epilepsy surgery. Epilepsy & Behavior. 2017;66:57–63. 10.1016/j.yebeh.2016.11.037 [DOI] [PubMed] [Google Scholar]
  • 25. Tanriverdi T, Dudley RWR, Hasan A, Jishi AA, Al Hinai Q, Poulin N, et al. Memory outcome after temporal lobe epilepsy surgery: Corticoamygdalohippocampectomy versus selective amygdalohippocampectomy. Journal of Neurosurgery. 2010;113(6):1164–1175. 10.3171/2009.10.JNS09677 [DOI] [PubMed] [Google Scholar]
  • 26. Mathon B, Bielle F, Samson S, Plaisant O, Dupont S, Bertrand A, et al. Predictive factors of long‐term outcomes of surgery for mesial temporal lobe epilepsy associated with hippocampal sclerosis. Epilepsia. 2017;58(8):1473–1485. 10.1111/epi.13831 [DOI] [PubMed] [Google Scholar]
  • 27. Hulse K, Grannan B, Kang J, Kim M, Tsai J, Hakimian S, et al. Non‐verbal memory decline is greater following anterior temporal lobectomy than mesial temporal laser ablation in non‐dominant temporal lobe epilepsy. Neurosurgery. 2023;69:110. 10.1227/neu.0000000000002375_500 [DOI] [Google Scholar]
  • 28. Powell HWR, Richardson MP, Symms MR, Boulby PA, Thompson PJ, Duncan JS, et al. Preoperative fMRI predicts memory decline following anterior temporal lobe resection. Journal of Neurology, Neurosurgery & Psychiatry. 2008;79(6):686–693. 10.1136/jnnp.2007.115139 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29. Baxendale S, Wilson SJ, Baker GA, Barr W, Helmstaedter C, Hermann BP, et al. Indications and expectations for neuropsychological assessment in epilepsy surgery in children and adults. Report of the ILAE neuropsychology task force diagnostic methods commission: 2017‐2021 neuropsychological assessment in epilepsy surgery. Epileptic Disorders. 2019;21:221–234. 10.1111/epi.16309 [DOI] [PubMed] [Google Scholar]
  • 30. Benedict RHB. Brief Visuospatial Memory Test‐Revised (BVMT‐R): Professional Manual. Lutz, FL: Psychological Assessment Resources; 1996. [Google Scholar]
  • 31. National Institute of Neurological Disorders and Stroke . Epilepsy common data elements: Recommended neuropsychology instruments–Adult (16+ years old). 2014, https://www.commondataelements.ninds.nih.gov/sites/nindscde/files/Doc/EPI/F1141_Recommended_Neuropsychology_Instruments_Adult.docx
  • 32. Schmidt M. Rey Auditory Verbal Learning Test. Los Angeles, CA: Western Psychological Services; 1996. [Google Scholar]
  • 33. Benedict RHB, Schretlen D, Groninger L, Dobraski M, Sphritz B. Revision of the brief visuospatial memory test: studies of normal performance, reliability, and validity. Psychological Assessment. 1996;8(2):145–153. 10.1037/1040-3590.8.2.145 [DOI] [Google Scholar]
  • 34. Landry A, Rouleau I, Colucci E, Nguyen DK, Boucher O. Adaptation of the Rey auditory verbal learning test and logical memory subtest from the Wechsler memory scales–3rd edition to assess accelerated long‐term forgetting in adults with epilepsy. Epileptic Disorders. 2025;27:1187–1200. 10.1002/epd2.70084 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35. Khalil AF, Iwasaki M, Nishio Y, Jin K, Nakasato N, Tominaga T. Verbal dominant memory impairment and low risk for post‐operative memory worsening in both left and right temporal lobe epilepsy associated with hippocampal sclerosis. Neurologia Medico‐Chirurgica. 2016;56(11):716–723. 10.2176/nmc.oa.2016-0004 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36. Tallarita GM, Parente A, Giovagnoli AR. The visuospatial pattern of temporal lobe epilepsy. Epilepsy & Behavior. 2019;101:106582. 10.1016/j.yebeh.2019.106582 [DOI] [PubMed] [Google Scholar]
  • 37. Gárgaro AC, Sakamoto AC, Bianchin MM, Geraldi C, Scorsi‐Rosset S, Coimbra ÉR, et al. Atypical neuropsychological profiles and cognitive outcome in mesial temporal lobe epilepsy. Epilepsy & Behavior. 2013;27(3):461–469. 10.1016/j.yebeh.2013.03.002 [DOI] [PubMed] [Google Scholar]
  • 38. Figueiredo P, Santana I, Teixeira J, Cunha C, Machado E, Sales F, et al. Adaptive visual memory reorganization in right medial temporal lobe epilepsy. Epilepsia. 2008;49(8):1395–1408. 10.1111/j.1528-1167.2008.01629.x [DOI] [PubMed] [Google Scholar]
  • 39. Gregory AM, Nenert R, Allendorfer JB, Martin R, Kana RK, Szaflarski JP. The effect of medial temporal lobe epilepsy on visual memory encoding. Epilepsy & Behavior. 2015;46:173–184. 10.1016/j.yebeh.2015.03.006 [DOI] [PubMed] [Google Scholar]
  • 40. Camina E, Güell F. The neuroanatomical, neurophysiological and psychological basis of memory: current models and their origins. Frontiers in Pharmacology. 2017;8:438. 10.3389/fphar.2017.00438 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41. Branco DM, Suarez RO, Whalen S, O'Shea JP, Nelson AP, da Costa JC, et al. Functional MRI of memory in the hippocampus: laterality indices may be more meaningful if calculated from whole voxel distributions. NeuroImage. 2006;32(2):592–602. 10.1016/j.neuroimage.2006.04.201 [DOI] [PubMed] [Google Scholar]
  • 42. Vannest J, Szaflarski JP, Privitera MD, Schefft BK, Holland SK. Medial temporal fMRI activation reflects memory lateralization and memory performance in patients with epilepsy. Epilepsy & Behavior. 2008;12(3):410–418. 10.1016/j.yebeh.2007.11.012 [DOI] [PubMed] [Google Scholar]
  • 43. Punia V. The landscape of epilepsy surgery in older adults: challenges and opportunities. Epilepsia Open. 2025;10(4):1009–1022. 10.1002/epi4.70062 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44. Hammond JB, Roth RM, Trask CL, Moncrief GG. Item and associative visual memory in presurgical temporal lobe epilepsy patients. International Neuropsychological Society. 2023;29(1):30–31. 10.1017/S1355617723001133 [DOI] [Google Scholar]
  • 45. Leeman‐Markowski BA, Martin SP, Hardstone R, Tam DM, Devinsky O, Meador KJ. Novelty preference assessed by eye tracking: a sensitive measure of impaired recognition memory in epilepsy. Epilepsy & Behavior. 2024;155:109749. 10.1016/j.yebeh.2024.109749 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46. Conradi N, Behrens M, Kannemann T, Merkel N, Strzelczyk A, Reif PS, et al. Factorial validity of a neuropsychological test battery and its ability to discern temporal lobe epilepsy from frontal lobe epilepsy: a retrospective study. Seizure. 2020;74:81–88. 10.1016/j.seizure.2019.11.012 [DOI] [PubMed] [Google Scholar]
  • 47. Brown FC, Tuttle E, Westerveld M, Ferraro FR, Chmielowiec T, Vandemore M, et al. Visual memory in patients after anterior right temporal lobectomy and adult normative data for the Brown location test. Epilepsy. 2010;17(2):215–220. 10.1016/j.yebeh.2009.11.026 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48. Landry A, Rouleau I, Desrochers V, Colucci E, Gravel V, Nguyen DK, et al. Novel analogous tasks to assess material‐specific memory impairments associated with temporal lobe epilepsy. Epilepsy & Behavior. 2025;166:110366. 10.1016/j.yebeh.2025.110366 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Table S1. Comparisons between the patients included in the final study sample and those excluded due to missing presurgical or post‐surgical BVMT‐R data or invalid performance.

Table S2. Distribution of pathology types and surgical interventions according to location and laterality of surgery.

Table S3. Pearson correlations between potential predictors and change in performance from pre‐ to postoperative assessment for the four main memory outcomes for the overall sample (N = 75).

Table S4. Association between predictors and change in performance after epilepsy surgery for mesiotemporal epilepsy.

EPD2-28-1175-s002.docx (35.8KB, docx)

Appendix S1.

EPD2-28-1175-s001.docx (11.3KB, docx)

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Articles from Epileptic Disorders are provided here courtesy of Wiley

RESOURCES