Verbal learning and forgetting in patients with temporal lobe epilepsy

Studies comparing non-surgical patients with left or right temporal lobe epilepsy (TLE) have shown irregular differences in verbal learning and memory. We assessed the performance of unoperated patients with epileptogenic temporal lobe lesions or cryptogenic TLE using a selective reminding procedure for the learning of a word list, and five delayed trials for the recall of learned words. On the selective reminding procedure, patients with left TLE were found to be more impaired than those with right TLE and controls, in agreement with the role of the left temporal lobe in verbal learning. The patients with right TLE were more impaired than the controls, possibly due to the semantic organization of the word list The rate of forgetting learned words was similar in the patient and control groups, suggesting that patients with left TLE can normally retain and/or retrieve stored items. These data support the hypothesis that distinct functional systems subserve learning and memory. Comparisons of the patient subgroups with epileptogenic lesions (hippocampal sclerosis or low-grade glioma) and those with cryptogenic TLE did not reveal any significant difference in learning or in memory, suggesting that epileptiform activity could affect verbal performance as a detectable temporal lesion.     

Preserved memory in temporal lobe epilepsy patients after surgery for low-grade tumour. A pilot study

The objective was to carry out a pilot study exploring memory outcome in patients with temporal lobe epilepsy (TLE) and low-grade tumour. A prospective study using a competence-related memory assessment was carried out in the Laboratory of Neuropsychology, Epilepsy Center and Neurosurgical Department of the “C. Besta” National Neurological Institute in 24 TLE patients undergoing surgical resection for left (n=12) or right (n=12) low-grade tumours and 36 healthy subjects. Patients underwent mesial or lateral temporal lobe lesionectomy. Neuropsychological tests exploring verbal and visual short-term memory, learning, delayed recall and ability to control interference in memory were applied. Before and after surgery, significant verbal impairment was present in left TLE patients compared to controls and right TLE patients, and visual deficits were present in both groups compared to controls. After surgery, there was no significant decrease in mean verbal or visual memory scores related to the operated side. Some memory abilities subserved by the contralateral temporal lobe improved. Postoperative memory scores were related to preoperative scores, side of operation, age and education. In patients with TLE and low-grade tumour, temporal lobe surgery does not necessarily induce memory deficits. Improvement of memory abilities subserved by the unoperated temporal lobe may be expected.     

The localization and lateralization of memory deficits in children with temporal lobe epilepsy

PURPOSE: It is often reported that children with temporal lobe epilepsy (TLE) experience nonlateralized memory impairments. However, many of these studies have been exploratory and not based on memory theory. Further, differences between mesial and lateral subgroups have not been adequately examined. This study aimed to discern more specific patterns of memory impairment in children with TLE. METHODS: Forty-three children (5-16 years) with lesional TLE participated. Subjects were categorized in terms of lesion laterality (left, n = 21; right, n = 22) and intratemporal location (mesial, n = 31; lateral, n = 12). Verbal and nonverbal memory tasks were administered that reflected associative, allocentric and recognition paradigms. RESULTS: Facial recognition was poorer in right TLE (p = 0.03). There were no differences between left and right groups on any other memory task, even when comparisons were restricted to cases with mesial involvement. Irrespective of laterality, clear differences were observed between mesial and lateral lesion subgroups (arbitrary associative learning, p = 0.01; complex figure recall, p = 0.03). The lateral lesion subgroup displayed intact memory function relative to normative standards. CONCLUSIONS: Memory is more frequently impaired in children with mesial as opposed to lateral TLE. Tasks with an associative component discriminated between these subgroups, supporting an associative model of hippocampal function. With the exception of facial recognition, memory deficits were not lateralized. Therefore, the nature of memory impairment experienced by children with TLE cannot be extrapolated from adult models.     

The Story Learning and Memory (SLAM) test: equivalence of three forms and sensitivity to left temporal lobe dysfunction

Remembering meaningful information is an important component of verbal memory. However, findings from existing story memory tests have been mixed in patients with temporal lobe epilepsy (TLE). We developed a test, the Story Learning and Memory (SLAM) test, in which a story is presented repeatedly until a performance criterion is reached, and verbatim recall is obtained only once, after a delay. In Study 1 we demonstrated a significant learning deficit in patients with left, but not right, TLE, and they were further impaired in retention of the story despite having learned it to the same criterion as subjects with right TLE and healthy subjects. These deficits remained confined to patients with left TLE after surgery. For clinical use we developed the SLAM in three versions in two languages; in studies 2 and 3 we tested and proved their equivalence.     

Semantic memory in partial epilepsy: verbal and non-verbal deficits and neuroanatomical relationships

Semantic memory was evaluated in 124 epilepsy patients, including 84 with left (n=44) or right temporal lobe epilepsy (TLE) (n=40) and 40 with left (n=25) or right frontal lobe epilepsy (FLE) (n=15), in order to determine their verbal and visual deficits, and the neuroanatomical relationships between them. The controls were 35 healthy subjects. Semantic memory was assessed by means of Picture Naming, Picture Pointing, the verbal Pyramid and Palm Trees Test (PPTT), the visual PPTT, Object Decision Hard, and Drawing From Memory. Episodic memory was assessed by means of the Short Story, Rey's Complex Figure, the Verbal and Visual Selective Reminding Procedure and Brown-Peterson Procedure. Factor analysis of the epilepsy patients distinguished their semantic memory scores from other neuropsychological domains. The semantic memory factor was significantly related to the side of the epileptic region, with lower scores in the left hemisphere and left TLE patients. In comparison with the controls, the left TLE patients were significantly impaired on Picture Naming, Picture Pointing, and Object Decision Hard. Subsequent analyses showed that, in comparison with the controls and the right TLE patients, the left TLE patients with lateral temporal lobe lesions were impaired in Picture Naming whereas, in comparison with the controls, the left TLE patients with mesial temporal lobe lesions were impaired in Object Decision Hard. On the contrary, the episodic memory factor was not related to the side of the epileptic region, and a few material-specific tests revealed opposite impairments in the left and right hemisphere patients. These results show that left TLE may cause semantic memory deficits involving verbal and visual information. Unlike the material-specific pattern of episodic memory, this pattern of impairment is in line with the view of an amodal semantic store in which all of the information about a thing overlaps. The semantic memory impairment may reflect damage in the lateral and mesial temporal lobe regions that impair neocortical functions in storing and retrieving information or hippocampal functions in processing meaningful stimuli.     

Compromised memory function in schizophrenia and temporal lobe epilepsy

The purpose of this study was to investigate compromised memory function of schizophrenia patients in comparison with temporal lobe epilepsy patients, whose memory impairments result from a clear lesion. The authors hypothesized that schizophrenia patients would show poorer immediate and delayed recall performances in verbal and visual memory tasks. The study sample consisted of a healthy comparison group of 30 subjects and three patient groups comprising 76 schizophrenia patients, 93 left temporal lobe epilepsy patients, and 72 right temporal lobe epilepsy patients. The authors assessed immediate recall, delayed recall, and delayed retention. Tasks were subdivided into two categories (easy and difficult), and then patient memory dysfunction was compared among the memory tests. The authors observed material-specific memory impairment, where the left temporal lobe epilepsy group showed severe verbal memory impairment and the right temporal lobe epilepsy group showed severe visual memory impairment. A moderate impairment was found in immediate and delayed verbal memory in schizophrenia patients, and the impairment of visual memory was amplified with delayed recall. Such a result can be interpreted not only as a generalized cognitive deficit, but also as an integrative dysfunction involving the mesial temporal and frontal lobes in the left and right hemispheres, whereby the lesion site cannot be determined selectively. Our results show that the selection of a memory task that cannot be influenced by verbal mediation is very important for analyzing memory dysfunction in schizophrenia patients.     

Adequacy of language function and verbal memory performance in unilateral temporal lobe epilepsy.

We examined adequacy of language functions, their influence on verbal learning and memory performance, and the relative effects of language function and laterality of seizure focus on the memory performance of 99 left-hemisphere dominant patients with invasively verified epilepsy of left (N = 47) or right (N = 52) temporal lobe origin. Patients with left temporal lobe epilepsy (TLE) scored significantly lower than the right TLE group on several aphasia battery subtests (Visual Naming, Sentence Repetition, Token Test, Reading Comprehension, Aural Comprehension). Adequacy of language function (nominal speech) was significantly related to verbal learning and memory performance for both left and right TLE groups. Finally, comparison of the predictive significance of laterality of TLE and adequacy of language function indicated that language functions (Visual Naming and Aural Comprehension), but not laterality of TLE, were significant predictors for verbal learning and memory performance. It is concluded that: 1) adequacy of basic language functions is particularly compromised in left TLE, 2) there is a significant relationship between adequacy of language function and several aspects of verbal learning and memory ability in both left and right temporal lobe groups, and 3) clinical assessment and theoretical models of memory need to consider these relationships.     

Memory in children with temporal lobe epilepsy is at least partially explained by executive dysfunction

An association between memory and executive dysfunction (ED) has been demonstrated in patients with mixed neurological disorders. We aimed to investigate the impact of ED in memory tasks of children with temporal lobe epilepsy (TLE). We evaluated 36 children with TLE and 28 controls with tests for memory, learning, attention, mental flexibility, and mental tracking. Data analysis was composed of comparison between patients and controls in memory and executive function; correlation between memory and executive function tests; and comparison between patients with mild and severe ED in memory tests. Children with TLE had worse performance in focused attention, immediate and delayed recall, phonological memory, mental tracking, planning, and abstraction. Planning, abstraction, and mental tracking were correlated with visual and verbal memory. Children with severe ED had worse performance in verbal and visual memory and learning tests. This study showed that ED was related to memory performance in children with TLE.     

Pseudoword and Real Word Memory in Unilateral Temporal Lobe Epilepsy

Recognition memory for pronounceable pseudowords (PWs), real words, and degraded photographs of unfamiliar faces, was examined in 45 patients with unilateral temporal lobe epilepsy (TLE), before and after Anterior Temporal Lobectomy, to test predictions from two accounts of hemispheric differences in memory functioning: (1) The ‘material-specific’ account predicts that left TLE (LTLE) patients would be impaired in memory for both familiar (real words) and unfamiliar (PWs) verbal stimuli, while memory for novel complex visual stimuli (unfamiliar faces) would be impaired in right TLE (RTLE) patients. (2) The ‘familiarity’ account predicts that memory for familiar stimuli (such as words) will be impaired in LTLE patients, while memory for both linguistic and nonlinguistic unfamiliar stimuli should be disrupted in RTLE patients. Results were consistent with the ‘material-specific’ hypothesis suggesting that both familiar and unfamiliar linguistic stimuli are processed for memory in the left medial temporal lobe (MTL), whereas unfamiliar nonverbal stimuli are processed for memory in the right MTL.     

Pseudoword and real word memory in unilateral temporal lobe epilepsy

Recognition memory for pronounceable pseudowords (PWs), real words, and degraded photographs of unfamiliar faces, was examined in 45 patients with unilateral temporal lobe epilepsy (TLE), before and after Anterior Temporal Lobectomy, to test predictions from two accounts of hemispheric differences in memory functioning: (1) The 'material-specific' account predicts that left TLE (LTLE) patients would be impaired in memory for both familiar (real words) and unfamiliar (PWs) verbal stimuli, while memory for novel complex visual stimuli (unfamiliar faces) would be impaired in right TLE (RTLE) patients. (2) The 'familiarity' account predicts that memory for familiar stimuli (such as words) will be impaired in LTLE patients, while memory for both linguistic and nonlinguistic unfamiliar stimuli should be disrupted in RTLE patients. Results were consistent with the 'material-specific' hypothesis suggesting that both familiar and unfamiliar linguistic stimuli are processed for memory in the left medial temporal lobe (MTL), whereas unfamiliar nonverbal stimuli are processed for memory in the right MTL.     

Effects of localised cerebral lesions and dysphasia on verbal memory.

Twenty-nine patients with unilateral left hemisphere lesions, 22 patients with unilateral right hemisphere lesions, and 19 neurological control patients with extracerebral lesions were assessed on verbal memory recall and recognition tests and on a battery of language tests. The left hemisphere group was significantly impaired in memory and language skills. Significant verbal memory impairment was found both in the subgroup of left hemisphere lesion patients whose lesions involved the temporal lobe and in the subgroup whose lesions did not. However, no significant differences between these left hemisphere subgroups' levels of performance on memory tasks emerged, even when dysphasia was taken into account. This study, therefore, fails to support the notion of a specific anatomical correlate of verbal memory impairment within the left hemisphere. Dysphasic subjects were significantly impaired on verbal memory tasks but displayed the same pattern of sensitivity to the effects of word frequency and word concreteness on verbal memory as control subjects, suggesting that the verbal memory of the dysphasic subjects was quantitatively rather than qualitatively impaired. This impairment could not be attributed to deficits in the comprehension or expression of the memory test items, and it is, therefore, proposed that language disturbances may hinder the efficient use of such language based procedures as may subserve verbal memory.     

Material-specific lateralization of working memory in the medial temporal lobe

Mnemonic deficits in patients with medial temporal lobe (MTL) damage arising from temporal lobe epilepsy (TLE) are traditionally constrained to long-term episodic memory, sparing short-term and working memory (WM). This view of WM as being independent of MTL structures has recently been challenged by a small number of patient and neuroimaging studies, which have focused primarily on visual and visuospatial WM. In the present study we investigated material-specific lateralization of WM in 96 patients with unilateral damage to MTL stemming from TLE (56 left) and 30 control subjects using a pair of matched verbal and visuospatial supraspan tasks. Patients with unilateral TLE were impaired on both verbal and visuospatial WM tasks irrespective of the affected hemisphere. Patients with unilateral right TLE showed an additional deficit for visuospatial WM capacity when contrasted with patients with left TLE, whereas patients with unilateral left TLE showed increased intrusion errors on the verbal task when compared to patients with right TLE. These findings suggest a material-specific lateralization of WM in the MTL.     

Depressed mood and memory impairment in temporal lobe epilepsy as a function of focus lateralization and localization

OBJECTIVE: Memory disorders and depressed mood are prominent psychological symptoms of temporal lobe epilepsies (TLEs). We examined the interaction of depressive mood and memory as a function of focus localization. METHODS: One hundred fifty-two TLE patients with right mesial (n=37, RTLE-AHS), right lateral (n=31, RTLE-LAT), left mesial (n=42, LTLE-AHS), and left lateral (n=42, LTLE-LAT) lesions and epilepsies underwent comprehensive presurgical evaluation and neuropsychological assessment of mood and memory. Univariate and multivariate analyses of covariance (ANCOVAs) and partial correlation analyses were performed to reveal interactions of depression and memory as a function of focus localization. RESULTS: No differences between the study groups were revealed for depression, indicating a general risk of 30% for depressed mood (BDI>12) in patients with TLE. ANCOVAs revealed significant main effects of focus side (left: verbal learning deficits; right: figural learning deficits) and site (mesial at disadvantage) on learning and memory scores. Correlation analyses revealed interactions between memory and mood only in LTLE-LAT patients. CONCLUSIONS: Although the data provide evidence that side and site of the epileptogenic region differentially affect material specific memory performance, there was no evidence of a specific temporal target region for depressive mood. In the majority of the patients, depressed mood and memory impairment appeared as independent rather than as related symptoms of TLE. In LTLE-LAT, however, mood was significantly related to verbal and figural memory performance. Epilepsy-driven pathological left temporofrontal circuits are discussed as a prerequisite for the coupling of mood disorders and memory impairment in this specific patient subgroup which is also known from the psychiatric major depression syndrome.     

Specificity of memory deficits after right or left temporal lobectomy

An impairment of verbal memory has consistently been associated with resection of the left dominant temporal lobe, whereas non-verbal memory deficits have been less reliably observed following resection of the right temporal lobe. Such a dissociation may be due to material-specific differences of processing between verbal and non-verbal information. Alternatively, the influence of the left and right limbic structures may vary according to the stage of memory processing. The aim of the study was to test these hypotheses by comparing verbal and spatial learning in patients with left or right temporal lobe resection for intractable epilepsy, using verbal and visuospatial memory tasks with the same design: control of encoding, multiple trial learning, free and cued recall, short and long delays. The results showed: (1) a similar pattern of learning and recall in the two groups; (2) a higher performance in spatial learning for patients with left temporal lobe resection and in verbal learning for patients with right temporal lobe resection; (3) material-specific effects characterized by a higher sensitivity to cues in the verbal domain and a better retention of information during delays in the spatial domain. These results suggest parallel processing of the two temporal lobes at the various memory stages, rather than an interaction between memory stage and side of the lesion similar to that already proposed for the frontal lobes. They also confirm a double dissociation between verbal/spatial information processing and side of temporal lobe resection.     

Accelerated long-term forgetting in temporal lobe epilepsy: Verbal,nonverbal and autobiographical memory

Patients with temporal lobe epilepsy (TLE) often present with memory complaints despite performing within normal limits on standard memory tests. One possible explanation for this phenomenon is accelerated long-term forgetting (ALF). The present study investigated material‐specific ALF in patients with unilateral TLE and also examined whether ALF could be demonstrated on a novel, standardized anterograde autobiographical memory (ABM) task. Fourteen patients with TLE and 17 controls were administered verbal, nonverbal and ABM event memory tasks. The participants were tested for immediate recall, recall and recognition at 30‐minute delay, and recall and recognition after four weeks. The extent of ALF was calculated based on the percentage decay of memory from the 30‐minute delay trial to the four-week delay trial. Patients with left TLE showed significantly greater ALF for verbal material and a trend towards greater forgetting of ABM. Patients with right TLE showed a non-significant trend towards greater ALF for nonverbal material. Patients with unilateral hippocampal abnormalities showed greater ALF compared to patients without hippocampal abnormalities. Patients with seizures that generalize had more global memory deficits and greater ALF. We conclude that patients with unilateral TLE show material‐specific ALF, which appears to be more pronounced with an abnormal hippocampus or seizures that secondarily generalize.     

Bilateral hippocampal atrophy: consequences to verbal memory following temporal lobectomy

BACKGROUND: Bilateral hippocampal damage is a risk factor for memory decline after anterior temporal lobectomy (ATL). OBJECTIVE: To investigate verbal memory outcome in patients with temporal lobe epilepsy (TLE) with either unilateral or bilateral hippocampal atrophy as measured by MRI. METHODS: The authors selected 60 patients with TLE who had undergone ATL (left = 31, right = 29). They determined normalized MRI hippocampal volumes by cursor tracing 1.5-mm slices from three-dimensional MRI acquisition. Hippocampal volumes were defined as atrophic if the volumes were below 2 SD for control subjects. Bilateral hippocampal atrophy was present in 10 patients with left TLE and 11 patients with right TLE. The authors assessed acquisition, retrieval, and recognition components of verbal memory both before and after ATL. RESULTS: Groups did not differ across age, education, intelligence, age at seizure onset, or seizure duration. Seizure-free rates after ATL were 70% or higher for all groups. Before surgery, patients with left TLE displayed worse verbal acquisition performance compared with patients with right TLE. Patients with left TLE with bilateral hippocampal volume loss displayed the lowest performance across all three memory components. After surgery, both groups of patients with left TLE exhibited worse verbal memory outcome compared with patients with right TLE. Bilateral hippocampal atrophy did not worsen outcome in the patients with right TLE. A higher proportion of patients with left TLE with bilateral hippocampal atrophy experienced memory decline compared with the other TLE groups. CONCLUSION: Bilateral hippocampal atrophy in the presence of left TLE is associated with worse verbal memory before and after ATL compared with patients with unilateral hippocampal volume loss or right TLE with bilateral hippocampal volume loss.     

Validation of the Wechsler Memory Scale-III in a population of people with intractable temporal lobe epilepsy

The Wechsler Memory Scale-Third Edition (WMS-III) was developed to overcome the limitations of the Wechsler Memory Scale-Revised and is now routinely used for pre-surgical assessment of memory for patients considering elective surgery for the relief of temporal lobe epilepsy (TLE). We investigated the validity of this relatively new measure in a population of people with resistant TLE.The sample consisted of 99 patients with a diagnosis if TLE in which the epileptogenic focus was clearly identified and localised to either the right or left hemisphere. Patients underwent a full neuropsychological assessment as part of their pre-surgical investigation, including the WMS-III.Patients with right temporal focus had significantly lower scores on the visual immediate and delayed indices of the WMS-III than they did on the corresponding auditory indices. The left temporal focal epilepsy group, however, showed no significant disparity between auditory and visual scores. The overall scores of the WMS-III were significantly lower than the normative data from the WMS-III standardisation samples.Like its predecessor the Wechsler Memory Scale-Revised, the WMS-III has limited value in identifying particular memory deficits associated with either left and right temporal lobe focus. The results of the WMS-III show that it is capable of lateralising to right hemispheric impairment but is more problematic in the assessment of left hemispheric impairment. The visual verbal discrepancy has questionable validity. The search for reliable and valid measures to distinguish between left and right temporal epileptic focus continues.     

Relation of sorting impairment to hippocampal damage in temporal lobe epilepsy

One hundred and twelve patients with left (n=65) or right (n=47) temporal lobe epilepsy (TLE), associated with mesial or lateral temporal lobe lesion, were compared to 53 patients with left (n=30) or right (n=23) frontal lobe epilepsy (FLE), in order to explore the contributions of hippocampal lesions and of memory deficits to sorting impairment. Thirty-six healthy subjects of similar age and education were controls. The Modified Wisconsin Card Sorting Test (MWCST) was used to explore sorting ability. The two-syllable word span and consistent long-term retrieval from the selective reminding procedure for word-list learning were used to evaluate memory. Raven's Coloured Progressive Matrices and Attentive Matrices served to control for abstract reasoning and attention. Left FLE patients and TLE patients with left hippocampal sclerosis were significantly impaired on MWCST, short-term memory, and word learning. TLE patients with other left hippocampal lesions were also impaired on MWCST, although not significantly so. Analysis of individual scores showed that 42% of TLE patients with left hippocampal sclerosis, 14% of TLE patients with other hippocampal lesions, 63% of left FLE patients, and 30% of right FLE patients were impaired on the MWCST. In patients with left hippocampal sclerosis, MWCST score was associated with the learning score provided by the selective reminding procedure and Raven's Coloured Progressive Matrices score, whereas in FLE patients, MWCST score was associated with Attentive Matrices score. These results suggest that only some TLE patients, i.e. those with hippocampal damage, may be expected to be impaired on card sorting. The impaired sorting ability of these TLE patients may be due to involvement of the hippocampal function in forming associations or in registering new information.     

Visual Learning on a Selective Reminding Procedure and Delayed Recall in Patients with Temporal Lobe Epilepsy

Summary: We administered a visual selective reminding procedure for the learning of designs to healthy controls and to left or right temporal lobe epilepsy (TLE) patients without demonstrable brain lesions. In comparison with the other groups, right TLE patients showed a deficit of learning characterized by an impairment in storing visual material and consistently retrieving it from long-term memory. Patients with left TLE performed similarly to controls. These data are consistent with the hypothesis of right hippocampal involvement in the learning of nonverbal material. Delayed retrieval of learned material was similar in all the subject groups tested at long time intervals, suggesting the impairment of learning in patients with right TLE does not necessarily imply a subsequent deficit in the retrieval of stored information. This finding may agree with the opinion that distinct functional systems subserve learning and memory.     

A critical evaluation of the lateralizing significance of material-specific memory deficits in patients with mesial temporal lobe epilepsy with hippocampal sclerosis

To critically assess the value of material-specific memory deficits in lateralizing temporal lobe dysfunction preoperatively, we compared the neuropsychological data of 50 consecutive patients with unilateral mesial temporal lobe epilepsy with hippocampal sclerosis (MTLE-HS; right: 31, left: 19) with those of 50 age- and education-matched healthy control subjects. On case-control comparison, both the subcohorts with left and right MTLE-HS performed poorly on intelligence tests, in addition to individual memory tests. However, comparison of the verbal and visual memory functions between subcohorts with right and left MTLE-HS revealed that learning trials and delayed word list recall were the only tests that hypothesized left temporal lobe dysfunction. We conclude that material-specific memory deficits are largely test driven, but there is a lateralizing role for task-specific deficits in left MTLE-HS. Although neuropsychological data help to define baseline neuropsychological impairment, caution should be exercised in interpreting the lateralizing value of material-specific memory deficits prior to surgery.