Autobiographical memory (AM) is understood as the retrieval of personal experiences that occurred in specific time and space. To date, there is no consensus on the role of medial temporal lobe structures in AM. Therefore, we investigated AM in medial temporal lobe epilepsy (TLE) patients. Twenty TLE patients candidates for surgical treatment, 10 right (RTLE) and 10 left (LTLE), and 20 healthy controls were examined with a version of the Autobiographical Interview adapted to Spanish language. Episodic and semantic AM were analyzed during five life periods through two conditions: recall and specific probe. AM scores were compared with clinical and cognitive data. TLE patients showed lower performance in episodic AM than healthy controls, being significantly worst in RTLE group and after specific probe. In relation to semantic AM, LTLE retrieved higher amount of total semantic details compared to controls during recall, but not after specific probe. No significant differences were found between RTLE and LTLE, but a trend towards poorer performance in RTLE group was found. TLE patients obtained lower scores for adolescence period memories after specific probe. Our findings support the idea that the right hippocampus would play a more important role in episodic retrieval than the left, regardless of a temporal gradient.
Cognitive neuroscience over the years has been trying to elucidate which are the basic mechanisms underlying autobiographical memory (AM). Despite the vast amount of studies performed in this area there is still no consensus on the role of medial temporal lobe (MTL) structures.
Medial temporal lobe epilepsy (TLE) patients provide a unique opportunity to systematically explore different aspects of AM processing considering the involvement of hippocampal structures on seizure onset and the connectivity to local and distal areas of MTL through the neural network related to epileptic spreading [
Two prominent theories argue the role of MTL in the encoding and retrieval of remote AM after consolidation. Briefly, the standard consolidation model (SCM) [
Most of the research in AM was conducted in aging, mild cognitive impairment, and degenerative disorders [
AM is understood as the retrieval of situations lived across lifetime and our own personal experiences that occurred in specific time and space [
Twenty patients with pharmacoresistant TLE candidates for surgery epilepsy were consecutively examined for this study at the Epilepsy Center, Neurology Department, Ramos Mejia Hospital of Buenos Aires, Argentina. Patients aged 18–53 years (
Demographic and clinical features.
TLE group | Controls | ||
---|---|---|---|
Left EZ | Right EZ | ||
|
10 | 10 | 20 |
Age (years) |
33,2 (10,97) | 30,3 (8,42) | 34,07 (11,47) |
Education (years) |
12,3 (3,09) | 12,4 (3,09) | 12,93 (2,96) |
Sex M/F | 7/3 | 5/5 | 12/8 |
Handedness | Right: 10 | Left: 2/right: 8 | Right: 20 |
MRI | HS = 9 |
HS = 8 |
NA |
Age at seizure onset (years) |
10,33 (7,69) | 8,2 (6,23) | NA |
Duration of epilepsy (years) |
23,44 (15,42) | 22,1 (11,79) | NA |
Seizure frequency (per month) | 8,57 (10,03) | 6,12 (6,93) | NA |
All subjects gave written informed consent approved by the Institutional Ethics Committee at Ramos Mejia Hospital, which follows the guidelines of the Declaration of Helsinki.
In order to determine lateralization and localization of the epileptogenic zone (EZ), video-EEG monitoring was performed in all patients over 5 days, finding 10 subjects with left EZ (LTLE) and 10 with right EZ (RTLE). An organized seizure activity with a clearly unilateral beginning was found in all patients, with a late propagation to contralateral areas only in 4 LTLE and 3 RTLE. Patients with a bilateral seizure activity from the beginning were not included. A magnetic resonance imaging (MRI) study was conducted for every patient: 19 subjects had hippocampal sclerosis, 10 right and 9 left, and one patient had a left temporal dysembryoplastic neuroepithelial tumor (Table
At the time of the study, all patients were polymedicated with 2-3 AEDs. Only one patient with LTLE had generalized tonic clonic seizures, while five patients (2 left and 3 right) had sporadic secondary generalized seizures.
A neuropsychological assessment was performed according to the CE presurgical protocol [
Neuropsychological performance of subjects (TLE).
Neuropsychological test | Left EZ | Right EZ | ||||
---|---|---|---|---|---|---|
Mean (SD) | Range | Mean (SD) | Range | |||
Min | Max | Min | Max | |||
WASI: |
||||||
Full Scale IQ (FIQ) | 94,66 (12,07) | 74 | 116 | 80,44 (9,44) | 70 | 96 |
Verbal IQ (VIQ) | 88,37 (14,69) | 70 | 107 | 78,11 (13,34) | 62 | 95 |
Performance IQ (PIQ) | 100,62 (15,9) | 73 | 126 | 84,22 (8,46) | 71 | 95 |
Verbal functioning |
||||||
RAVLT (delayed recall) | −1,36 (1,25) | −2,96 | 0,29 | −0,42 (0,67) | −1,52 | 0,56 |
BNT | −3,02 (3,17) | −7,78 | 1 | −2,2 (1,81) | −5,02 | 0,76 |
Verbal fluency (phonemic) | −0,90 (0,62) | −2 | −0,05 | −0,73 (0,9) | −1,68 | 1,1 |
Visual functioning |
||||||
RCFT (delayed recall) | −1,41 (1,63) | −3,52 | 0,56 | −1,56 (1,38) | −3,32 | 0,66 |
WASI: Wechsler Abbreviated Scale of Intelligence. RAVLT = Rey Auditory Verbal Learning Test. BNT: Boston Naming Test. Phonemic verbal fluency. RCFT: Rey Complex Figure Test.
AM was assessed with the Autobiographical Interview [
According to the administration instructions [
Every interview was recorded, transcribed, and segmented in detail or pieces of information. The details were classified as internal-episodic and external-semantic. These were further divided into the following categories: main event, place, time, perceptual details, and thoughts/emotions as internal-episodic information; and repetitions, other details (metacognitive, editorial statements), and factual information as external-semantic information. The information given was segmented and scored to obtain quantitative data following the scoring instructions [
Additionally, each person assigned a value between 1 and 6 related to how well they visualized the event related, the emotional change produced by the event, the importance given actually and then, and how frequently they talk or think about it. Quantitative ratings were also assigned for episodic information (time, place, perception, and thoughts/emotion) and time integration on a scale of 0 to 3 and episodic richness using a scale extended to 6 points [
AI scores were compared between all patients and control group, in RTLE/LTLE versus control, and between LTLE and RTLE. The clinical data and the cognitive status were also analyzed.
Control and TLE group were matched for age, sex, and formal education. For each patient, the raw values of every cognitive test in the neuropsychological battery were normalized to a
We compared TLE groups versus controls’ performance in AI and we analyzed the composite measures considering both the total life span and each period of time. One-way ANOVA, Student’s
All comparisons that were significant at the
The total number of episodic and semantic details recalled across five life periods was compared for TLE group and control subjects.
For episodic details, TLE group scores were lower during recall than control group but we did not find a statistical difference (
Mean number of episodic and semantic details retrieved during recall (darkest portion of the histogram) and specific probe (lighter portion of the histogram) conditions for each group: RTLE, LTLE, and control.
In relation to semantic details, no differences were found between TLE and controls during either recall (
The number of episodic and semantic details retrieved for individual categories was compared between TLE and control group.
For episodic details, differences were found between TLE group and control only after specific probe in event details (
Mean number of (a) episodic and (b) semantic details retrieved for each category during recall and specific probe conditions. All significance levels were at
For semantic details (Figure
Figure
Number of details and ratings retrieved across each one of the life periods, childhood (>10 years), adolescence (10–18 years), early adulthood (18–30 years), and adulthood (30–55 years), and during the last year, for recall (left column) and after specific probe (right column). (a) shows the average number of episodic details, (b) indicates performance for semantic details, and (c) indicates the total score in different rating categories (max = 21). Bars indicate the standard error of the mean (SEM).
Ratings composites for recall condition were significantly diminished for RTLE compared to controls not only for adolescence period (
Figure
Average of subjective quality ratings for each of the five time periods regarding the vividness of the event, the emotional change experienced, how important the event was at the time of its occurrence and now and how frequently they thought or spoke about it. Bars indicate the standard error of the mean (SEM).
No differences were found when participants were asked how important the event actually is, how relevant it was at the moment of its occurrence, and how frequently they rehearse about it.
AI scores were compared to neuropsychological test results for each patient. A statistically significant correlation was only found between VIQ and episodic details during recall in RTLE (Pearson correlation coefficient
There were no significant correlations between AI scores and age, years of education, age of onset, and disease duration. Our sample does not include cases of recent onset; the duration of the epilepsy was higher than ten years.
Most of previous memory research in epileptic subjects has analyzed different aspects related to type of material, lateralization, and pre- and postsurgical performance [
Different studies that have used the AI [
However, other authors suggest that the left hippocampus is essential to episodic retrieval [
In relation to episodic detail categories, RTLE presented impaired performance during retrieval after specific probe for perceptual category. No differences were found between RTLE and LTLE. St-Laurent et al. [
Our findings showed no temporal gradient through the five life periods in episodic and semantic AM, as proposed by the consolidation theory [
With regard to semantic AM memory we observed higher performance of LTLE patients on recall condition that disappears after specific probe. In the same way, higher scores in semantic categories, factual information (semantic), and metacognitive statements (the other category) were found which would reflect a compensatory cognitive mechanism [
In our work, participants showed no difference with respect to the degree of vividness, personal significance, and rehearsal of their recollections. We could assume that a lesion in MTL structures does not affect these subjects’ appreciation. As far as cognitive performance is concerned, we did not observe any relation between verbal or visual memory performance and AI scores that could provide additional data to our results.
Finally, regarding the duration of epilepsy, our interest was to determine its influence in AM recollections but none of the subjects had illness duration fewer than 10 years, so this variable could not be analyzed. The impact of epilepsy across life period recollections was not analyzed either, because the age of onset for the majority of subjects was during childhood.
Our results allow us to provide additional evidence to previous work, of the hippocampal structures involvement in episodic autobiographical memories recollection, particularly the right hippocampus in TLE patients. It is important to consider that one limitation of our study is the relatively small sample size that may contribute to the lack of differences between EZ side groups. For future work, we will compare performance in TLE patients, before and after surgery.
Temporal lobe epilepsy
Autobiographical memory
Autobiographical Interview.
The authors confirm that they have read the journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.
The authors declare that there is no conflict of interests regarding the publication of this paper.
The authors would like to thank Dr. Brian Levine for providing the Autobiographical Interview administration and scoring manual and for allowing them to translate it to Spanish, Lucia Valeriano for helping them with the transcriptions, and all the participants for giving their time. This study was supported by a doctoral fellowship from the National Scientific and Technical Research Council (CONICET) to Claudia P. Múnera.