Keywords
Seizures - Memory, Episodic - Neuropsychology - Spatial Processing
INTRODUCTION
The term autobiographical memory (AM) refers to a system encompassing memories encoded in the distant personal past.
Traditionally, AM is defined as a form of episodic memory for personally-experienced
events specific in time and place.[1] However, many researchers suggest that AM can be delineated into two intercorrelated
but separate subsystems: autobiographical episodic memory (AEM) and autobiographical
semantic memory (ASEM);[2] AEM pertains to specific and unique personal incidents, while ASEM relates to repeated
and generalized personal facts.[2] Here are examples that may be of some help for the reader: AEM –“I remember the
first day in Rome, when I missed the appointment with the house owner due to a flight
delay and I called her from the metro station”; ASEM – “When I was fifteen, I spent
Fridays practicing trumpet on specific military tracks and marching maneuvers for
upcoming marching band performances”. The functional differentiation of AM appears
to manifest at a neuroanatomical level as well. Studies investigating the neural underpinnings
of AM indicate that AEM tends to involve the right medial temporal lobe, while ASEM
seems to rely more on the left medial temporal lobe.[3]
Autobiographical memory in temporal lobe epilepsy patients
Autobiographical memory in temporal lobe epilepsy patients
In addition to research focusing on healthy populations, the study of AM impairment
could also be extended to patient populations, where it is as a significant cognitive
comorbidity in various psychiatric[4] and neurological conditions. Notably, AM impairment features prominently in conditions
such as epilepsy.[5] The neuropsychological exploration of memory function and dysfunction predominantly
focused on temporal lobe epilepsy[6] (TLE), since the temporo-limbic system and its associated cognitive implications,
rendering TLE a valuable tool for probing memory processes.
Several imaging, behavioral, and neuropsychological data support the role of the hippocampi
in episodic memory.[7] It is well established that memory deficits in TLE are partly linked to epileptogenic
lesions, such as hippocampal sclerosis,[8] as well as to the presence of a wider group of areas that are functionally abnormal
in the interictal period (referred to as the functional deficit zone).[9] Additionally, temporolimbic dysfunction in TLE has sparked debate among researchers
regarding the relationship between epileptic activity and its effects on memory.[2] Therefore, TLE patients appear to be ideal candidates for the study of seizure-induced
memory impairment, especially AM, due to the involvement of temporolimbic structures
in memory and learning.
Autobiographical memory in genetic generalized epilepsy patients
In contrast, AM study in other types of epilepsy has been largely neglected, particularly
in the case of genetic generalized epilepsies (GGEs), which are highly-heritable conditions,
possibly determined by polygenetic mechanisms, which account for 15% to 20% of all
adult cases of epilepsy.[10] They comprise four key epilepsy syndromes: childhood absence epilepsy (CAE), juvenile
absence epilepsy (JAE), juvenile myoclonic epilepsy (JME), and epilepsy with generalized
tonic-clonic seizures alone (EGTCA), each exhibiting different electroclinical characteristics[11] and neuropsychological profiles[12] compared to focal epilepsies. This makes the study of remote memory (AM) in GGEs
particularly intriguing, given the known effects of seizures on the consolidation
of newly-acquired material in long-term memory. However, knowledge of whether primary
generalized seizures per se link to deficits in the recall of AMs remains scarce.[13] The uniqueness of AEMs and their inability to be completely reestablished once compromised
may render them more susceptible to disruption than ASEMs in view of the diffuse brain
pathology and/or generalized seizures characterizing GGEs.
Compared to TLE patients, GGE patients show no morphological brain abnormalities on
magnetic resonance imaging (MRI) scans and often do not display clinical signs between
seizures.[14] Early neuropsychological studies into GGE suggested a primary impairment in sustaining
and directing attention,[15] along with diminished performance in attention, memory, and psychomotor speed.[16] Moreover, dysexecutive manifestations have been observed in GGE patients.[17]
[18]
To date, while cognitive profiling of GGEs has garnered attention, investigations
into AM in adult GGE patients remain scarce, with pediatric patients receiving slightly
more consideration.[13]. Given this gap, the present study aims to explore AM in adult GGE patients, specifically
focusing on patients with JME, JAE and EGTCA. Moreover, we aim to assess the potential
influence of seizures and disease-related factors on memory function,[13] in an attempt to discuss the putative role of cognitive pathophysiological mechanisms
underlying declarative memory (DM) deficits in GGEs. Here, the term cognitive pathophysiology refers to the pathophysiology of cognitive disorders.
Before presenting our findings, we have considered important to provide the reader
with a brief outline of neuropsychological impairments in GGEs, and how these impairments
may affect AM.
Neuropsychological differences among GGE subtypes
Furthermore, the question of whether cognition is differentially affected in GGE subtypes
remains a matter of speculation. In relation to the CAE and EGTCA subtypes, there
is a general consensus that JME patients exhibit characteristic features of cognitive
impairment, which may be genetically determined and distinct from drug effects and
seizure impacts.[19] Recent findings[20] point on the susceptibility of the auditory information processing in GGE patients
(EGTCA and JME included) to the effects of idiopathic generalized seizures. Moreover,
dysfunction in executive attention appears to be a salient neuropsychological feature
in JME.[21] The EGTCA patients face an increased risk of developing psychiatric and neurocognitive
comorbidities;[22] however the characterization of cognitive dysfunction patterns in this syndrome
remains an ongoing area of research.[19]
Additionally, there is evidence of visuospatial and language deficits in CAE and JAE,
which are associated with atypical neurodevelopment.[19] Reduced right hippocampal volume has been linked to JAE,[23] while altered hippocampal structure and functional reorganization have also been
observed in other GGE subtypes.[24] The increased state of brain excitability in the absence of seizures may contribute
to interictal functional deactivation of the hippocampus and parahippocampal gyrus,
possibly accounting for memory, reasoning, and executive deficits in JAE.[19] Finally, neuroimaging evidence[25] suggests that regions in the basal ganglia-thalamocortical network exhibit aberrant
nodal centrality. As epilepsy progresses, reorganization of the caudate nucleus may
aid in preventing seizure attacks, while changes in the hippocampus may serve to protect
against cognitive impairments.
Possible interactions regarding cognitive impairment in GGEs and AM
One might anticipate that AM performance could be linked to the aforementioned neuropsychological
abnormalities of GGEs. This expectation arises from substantial evidence[26] highlighting the crucial role attention plays in the encoding of episodic memories,
while internal attention (selection, modulation, and maintenance of internally-generated
information) not only facilitates the retrieval of episodic memories but is also indispensable
to render them conscious. When it comes to AM retrieval, voluntary retrieval activates
more prefrontal regions compared to involuntary retrieval,[27] since the former necessitates a cyclic search process heavily reliant on executive
control areas to regulate top-down processing.[28]
A pertinent issue that requires clarification pertains to deficits in remote memory,
which also encompasses AM, as a likely consequence of long-standing anterograde amnesia,
wherein neurological damage impairs memory accumulation. However, there is evidence[29] suggesting that retrograde amnesia, which characterized by the loss of, or loss
of access to, previously well-established memories, can also occur, albeit rarely,
in the absence of any anterograde memory impairment. Reports[29] of neurological diseases causing such “focal retrograde amnesia” have generated
considerable debate, an issue also relevant to GGEs.
METHODS
Participants
We used a case-control design for the present study. A convenience sample of 21 patients
with GGE (JAE, EGTCA, and JME; with 10 female subjects) participated voluntarily.
A unified GGE group was formed, comprising all three sub- syndromes (JME: n = 9; JAE:
n = 6; and EGTCA: n = 7), with a mean age of 28.19(± 11.70) years and a mean of 12.33(± 2.39)
years of schooling ([Table 1]). In the total GGE sample, 5 patients were seizure free, and 9 were on monotherapy.
Table 1
Demographic and clinical characteristics
|
Clinical characteristics
|
GGE patients (n = 21)
|
HCs (n = 21)
|
Test statistic (U-test and t-test where appropriate)
|
p-value
|
|
Sex (female): n
|
10
|
9
|
0.096a
|
0.757
|
|
Age (years): mean ± SD
|
28.19(± 11.70)
|
32.05 (9.72)
|
−1.789
|
0.074
|
|
Years of schooling: mean ± SD
|
12.33(± 2.39)
|
13.52 (1.94)
|
−1.366
|
0.172
|
|
Age at epilepsy onset (years): mean ± SD
|
16.11(± 6.99)
|
N/A
|
|
|
|
Yearly frequency of seizures: mean ± SD
|
2.25(± 1.48)
|
N/A
|
|
|
|
Duration of epilepsy (years): mean ± SD
|
13.05(± 13.90)
|
N/A
|
|
|
|
Seizure freedom (years): mean ± SD
|
3.43(± 4.60)
|
N/A
|
|
|
|
Number of pills (ASMs): mean ± SD
|
1.56(± 0.89)
|
N/A
|
|
|
Abbreviations: ASMs, antiseizure medications; GGE, genetic generalized epilepsy; HC,
healthy controls; N/A, not available; SD, standard deviation.
Note: Seizure freedom is the time without occurrence of seizures.
The study was approved by the Ethics Committee of the First Department of Neurology
of the American Hellenic Educational Progressive Association (AHEPA) University Hospital
of Thessaloniki. All patients were seen at the Epilepsy Outpatient Clinic of a Tertiary
Academic Hospital and provided informed consent for the procedures. The inclusion
criteria were patients diagnosed by a consulting neurologist based on clinical and
neurophysiological evidence, including MRI and video-electroencephalography (VEEG)
monitoring. The exclusion criteria were history of brain injury, posttraumatic epilepsy,
psychiatric and/or neurological disorders other than seizures, and mental retardation.
Moreover, all patients underwent a comprehensive clinical interview to gather detailed
seizure histories, including the precise onset and duration of seizures, the daily
intake of antiseizure medications (ASMs), seizure frequency, and periods of seizure
freedom.
Control group
Healthy volunteers were invited by the research team to participate in the control
group. The group of healthy controls (HCs) consisted of 21 participants (9 female
subjects), matched demographically in terms of age, years of schooling, and gender.
The HCs had an mean age of 32.05(± 9.72) years and a mean of 13.52(± 1.94) years of
schooling. The gender distribution, mean age, and years of schooling did not significantly
differ between the two groups ([Table 1]).
The exclusion criteria for the control group included a history of psychiatric illness,
head injury, and/or other neurological disorders. The HCs were screened by our expert
neurologists after a clinical interview. All participants were native Greek speakers,
predominantly residing in urban areas of Northern Greece.
Materials
Neuropsychological measure
The neuropsychological measure implemented in this study are briefly described bellow.
The Autobiographical Memory Interview[30]
The Autobiographical Memory Interview (AMI) is a semistructured interview schedule
to assessing personal remote memory. The interview consists of two types of questions:
personal semantic and autobiographical incidents, concurrently administered, that
is, autobiographical episodes embedded with semantic knowledge. Thus, by conducting two separate interviews for both episodic and semantic memory,
these latter AM memory components can be directly compared. In the episodic autobiographical
interview, the interviewee is asked to provide descriptions of incidents across three
broad periods: childhood (ages 0–18 years), early adulthood (ages 18–30 years), and
the recent past (within the past 5 years), with each period having a maximum score
of 21 points. Prompt questions raising the issue were asked about experiences with
a friend or teacher during elementary and secondary school, first day of work at a
job, wedding, a visit from a relative, a journey. The episodic richness and specificity
of time and place can be then analyzed by the examiner. After that, the personal semantics
interview aims at gathering general information about the personal past of the subject,
that is, names of friends or work locations, from the three different periods.[30] It assesses the temporal gradient for episodic and personal semantic memory, irrespective
of the life periods. The interview requires the examinee to recall memories of commonly-experienced
events, rather than to respond to word cues.[30]
The AMI was translated and adapted to the Greek language, and its administration and
scoring procedures followed the guidelines of the AMI manual. Due to the relatively
young age of the GGE patients and HCs, we decided to focus solely on two periods (childhood
and the recent past) from the AMI. This choice was made due to the considerable overlap
between early-adulthood and the recent past within our sample.
Personal semantic questions
The participants were required to recall facts from their past, such as home addresses,
the locations of attended schools, and friends' names.
Questions on autobiographical incidents
The participants were asked to describe life incidents that occurred during each of
the two periods. It is worth noting that we only included two out of the three AMI
periods in our analysis. In their narratives, subjects were asked to provide specific
temporal and spatial contextual information. For each period, three incidents were
probed, such as “final school excursion”, for example. The participants' replies were
recorded as verbatim as possible, and their memories were checked using information
from medical files and reports from relatives.
All questions on autobiographical incidents were rated by two independent researchers,
with the second being blinded to the subject group and the ratings given by the other
researcher. The correlations between the scores provided by the 2 researchers for
each period were r = 0.942 (childhood) and r = 0.948 (recent past). To test the authenticity and accuracy of the memories, relatives
were asked to confirm the patients' recollections.
Memory for stories
We used the immediate story recall condition from the Logical Memory (LM) subtest
of the Wechsler Memory Scale-Revised (WMS-R),[31] to assess immediate verbal episodic recall.
Visual short-term/immediate memory, visuoperceptual ability, and visuospatial memory
To assess visual recall, the Visual Patterns Test (VPT),[32] a measure of short-term nonverbal memory (visual working memory), was implemented
given its relative independence of any spatial memory components.
The Rey-Osterrieth Complex Figure Test (RCFT)[33] assesses perceptual organization and visual memory. We used only the RCFT-copy section
to assess visuoperceptual ability, and the RCFT delayed recall to assess visuospatial
memory.
Everyday memory
Everyday memory was assessed through the Greek version of the Rivermead Behavioral
Memory Test (RBMT),[31] which consists of several subtests that evaluate several memory functions, including
immediate and delayed recall, recognition, and semantic and prospective memories.
These subtests include RBMT name (delayed recall)/remembering people's names, RBMT
story/remembering stories (immediate and delayed recall), RBMT pictures/remembering
objects (delayed recognition of 10 out of 20 pictures), RBMT faces/remembering faces
(delayed recognition of 5 out of 10 faces), RBMT personal item/remembering to collect
personal belongings (delayed recall of the place where a subject's personal item has
been hidden) RBMT bell (prospective memory)/scheduled actions, RBMT orientation/remembering
a route (semantic memory, knowledge of personal or general facts), and RBMT date/remembering
an appointment (semantic memory, general knowledge questions).
Verbal executive functions and semantic memory
Verbal executive functions and semantic memory were evaluated through the phonemic
verbal fluency (PVF) test, which assesses spontaneous word production starting with
a given letter, while the semantic verbal fluency (SVF) variant entails the retrieval
of words within a specific conceptual category (such as “animals”, for example).[34]
Procedure
The neuropsychological assessment of both groups was conducted in two separated sessions
scheduled on the same day, with a break in between sessions.
Statistical analysis
Initially, the data were examined visually for extreme outliers, that is, observations
that are more extreme than the first quartile (Q1) - 3 x the interquartile range (IQR) or the third quartile (Q3) + 3 x the IQR), and all such data points were excluded from further analysis. Descriptive
statistics were then computed for the clinico-demographic and neuropsychological variables.
The assumption of normality of the data was tested using the Shapiro-Wilk test, since
it is more powerful for small sample sizes (n < 50) than the commonly used Kolmogorov-Smirnov
test.[35] When the normality hypothesis was rejected, the non-parametric Mann-Whitney U-test
was used to examine differences between the two groups; otherwise, the standard t-test was used.
For cases in which the normality assumption was met, Hedge's g, which is a transformation
of Cohen's d, was used as an effect size indicator. For cases in which the normality
assumption was not met, Cliff's delta was initially estimated and then transformed
to Cohen's d and Hedge's g to ensure the comparability of effect size indicators.
Moreover, correlation coefficients involving autobiographical memory, other neuropsychological
measures, and clinical variables were computed.
All statistical tests were conducted at a 5% significance level, and the statistical
analyses were performed using the IBM SPSS Statistics for Windows (IBM Corp., Armonk,
NY, United States) software, version 27.0, and R language (R Foundation for Statistical
Computing, Vienna, Austria) version 4.1.0.
RESULTS
The demographic and clinical characteristics of the sample are presented in [Table 1]. The GGE group, performed significantly worse compared to the controls in nearly
all neuropsychological measures administered ([Table 2]). Specifically, the former exhibited poorer performance in immediate (LM story Ire-WAIS,
RBMT story-ImR) and delayed episodic recall (RBMT story-DRe), visuospatial working
memory (VPT1), visuoperceptual organization (RCFT-copy), face recognition memory (RBMT
faces), and verbal executive functions (SVF, SVF-clusters, PVF). Additionally, GGE
patients showed impaired performance in both AEM (childhood and recent past) and ASEM
(childhood and recent pest) periods compared to to HCs ([Table 3]).
Table 2
Neuropsychological performance
|
HCs
|
GGE
|
Test statistic (U-test and t-test where appropriate)
|
p-value
|
g
|
|
Neuropsychological measures
|
|
LM story I Re-WMS-I
|
mean(± SD); (IQR)
|
18.29(± 3.84);
(16.0)
|
14.90(± 3.56);
(15.0)
|
2.960 (df = 39)
|
0.005
|
0.896
|
|
VPT1 visual
|
mean(± SD); (IQR)
|
23.33(± 2.82);
(11.0)
|
17.75(± 4.09);
(17.0)
|
5.113 (df = 39)
|
0.000
|
1.567
|
|
RCFT-copy
|
mean(± SD); (IQR)
|
35.90(± 0.44);
(2.00)
|
33.62(± 6.52);
(30.0)
|
305.5 (U)
|
0.004
|
0.596
|
|
RCFT-DRe
|
mean(± SD); (IQR)
|
21.14(± 6.24);
(25.0)
|
17.57(± 7.17);
(31.0)
|
1.721 (df = 40)
|
0.093
|
0.521
|
|
RBMT story-I ImRe
|
mean(± SD); (IQR)
|
19.14(± 3.32);
(11.0)
|
12.12(± 3.36);
(12.0)
|
309.0 (U)
|
0.000
|
2.128
|
|
RBMT story-I Dre
|
mean(± SD); (IQR)
|
8.90(± 0.30);
(1.0)
|
8.56(± 0.51);
(1.0)
|
225.5 (U)
|
0.019
|
0.512
|
|
RBMT pictures
|
mean(± SD); (IQR)
|
9.95(± 0.22);
(1.0)
|
9.88(± 0.34);
(1.0)
|
181.0 (U)
|
0.418
|
0.099
|
|
RBMT faces
|
mean(± SD); (IQR)
|
4.95(± 0.22);
(1.0)
|
4.62(± 0.50);
(1.0)
|
223.0 (U)
|
0.014
|
0.485
|
|
RBMT name
|
mean(± SD); (IQR)
|
3.71(± 0.72);
(2.0)
|
3.25(± 1.24);
(2.0)
|
198.0 (U)
|
0.206
|
0.241
|
|
RBMT personal items
|
mean(± SD); (IQR)
|
3.62(± 0.80);
(2.0)
|
2.75(± 1.61);
(4.0)
|
215.0 (U)
|
0.072
|
0.407
|
|
Semantic memory tests
|
|
SVF
|
mean(± SD); (IQR)
|
51.86(± 5.47);
(18.0)
|
37.50(± 8.52);
(37.0)
|
395.0 (U)
|
0.000
|
2.444
|
|
SVF-clusters
|
mean(± SD); (IQR)
|
7.14(± 2.2);
(9.0)
|
4.25(± 1.41);
(5.0)
|
372.5 (U)
|
0.000
|
1.762
|
|
PVF
|
mean(± SD); (IQR)
|
35.00(± 8.33);
(30.0)
|
22.77(± 6.81);
(23.0)
|
4.960 (U)
|
0.000
|
1.561
|
|
PVF-clusters
|
mean(± SD); (IQR)
|
1.72(± 1.38);
(5.0)
|
1.28(± 1.45);
(5.0)
|
232.0 (U)
|
0.211
|
0.317
|
Abbreviations: df, degrees of freedom; Dre, delayed recall; GGE, genetic generalized
epilepsy; HC, healthy control; ImRe, immediate recall; IQR, interquartile range; LM
story I Re-WMS-I, Logical Memory story immediate recall-Weschsler Memory Scale-I;
PVF, phonemic verbal fluency; RBMT story-I, RBMT-faces recall story I; RCFT, Rey-Osterrieth
Complex Figure Test; SD, standard deviation; SVF, semantic verbal fluency; VPT, Visual
Patterns test.
Table 3
Performance on the AMI
|
AMI measures
|
HCs
|
GGE
|
Test statistic (U-test)
|
p-value
|
g
|
|
AEM-childhood
|
mean(± SD);
(IQR)
|
8.71(± 0.72);
(3.0)
|
6.67(± 2.04);
(7.0)
|
353.5
|
0.000
|
1.382
|
|
AEM-recent past
|
mean(± SD);
(IQR)
|
8.57(± 0.68);
(2.0)
|
6.45(± 1.67);
(6.0)
|
370.0
|
0.000
|
1.705
|
|
ASEM-childhood
|
mean(± SD);
(IQR)
|
20.74(± 0.62);
(2.5)
|
18.77(± 1.72);
(5.5)
|
366.5
|
0.000
|
1.629
|
|
ASEM-recent past
|
mean(± SD);
(IQR)
|
20.48(± 0.98);
(4.0)
|
19.22(± 1.45);
(5.0)
|
330.5
|
0.001
|
1.035
|
Abbreviations: AEM, autobiographical episodic memory; AMI, Autobiographical Memory
Interview; ASEM, autobiographical semantic episodic memory; GGE, genetic generalized
epilepsy; HC, healthy control; IQR, interquartile range; SD, standard deviation.
No significant differences were found in terms of the performance in the AMI or other
cognitive tests (OCTs) between GGE patients receiving mono- and polytherapy. Moreover,
neither the age at onset nor the duration of epilepsy appeared to significantly affect
the cognitive performance of GGE patients.
All correlation coefficients regarding epilepsy duration, epilepsy onset, AMI, and
OCTs were weak, except for a significant negative correlation involving epilepsy duration
and the measures of RCFT-DRe and VPT. Regarding the correlations concerning AMI measures
and OCTs, AEM-childhood showed a significant positive correlation with RCFT-copy,
while AMI-recent exhibited a significant negative correlation with RBMT-name. Moreover,
ASEM-childhood demonstrated significant positive correlations with RCFT-copy and RBMT-pictures,
while no significant correlations were identified for ASEM-recent.
DISCUSSION
The present study aimed to investigate AM in a sample of adult GGE patients compared
to HCs, and tested for differential impairments in AEM and/or ASEM, as well as in
other domains, such as episodic, visual, and semantic memories, and executive function.
We found that GGE patients experienced overall impairments in retrieving autobiographical
episodic and semantic information for both periods compared to HCs.
Several studies have examined the impact of seizure-related variables on remote memory.
Evidence[36] suggests that factors such as epilepsy duration and history of generalized seizures
are unlikely to affect memory for autobiographical or public facts and events, a finding
consistent with our results.
Furthermore, GGE patients exhibited differences from HCs in the domain of visuoperceptual
organization (RCFT-copy). Of note, visuospatial impairments in JAE patients have been
linked to atypical neurodevelopment[19] and reduced right hippocampal volume.[23]
The diminished performance of GGE patients in the verbal executive domain (SVF, SVF-clusters,
PVF) and visual working memory (WM; Visual Pattern Task, VPT) aligns with previous
research highlighting executive function abnormalities in GGE, particularly in JME,
in which frontal dysfunction has been documented.[21]
[37] Hopefully, these findings may serve to provide insights into the cognitive pathophysiology
of GGEs.
In general, compared to HCs, the GGE group exhibited poorer performance in immediate
and delayed episodic recall measures (LM story IRe-WMS-I, RBMT story-ImRe and Dre,
[Table 2]), a finding partially consistent with recent findings.[20] Interestingly, interictal functional deactivation of the hippocampus and para-hippocampal
gyrus in JAEs may contribute to deficits in memory, reasoning, and executive function.[25]
[38] Recent neuroimaging evidence[25] further suggests that the limbic system plays a role in abnormal resting-state functional
networks in JAE. Similarly to the known effects of frontal lesions in episodic memory,[39] JME patients were expected to present with deficits in free episodic recall, given
their well-documented presence of frontal lobe dysfunction.[24]
[37] Regarding GGEs, impaired face recognition memory has been observed, as evidenced
by immediate face memory recognition problems in GEE patients.[40]
However, it remains uncertain whether this finding represents a primary memory disorder
or rather a secondary systemic manifestation due to primary seizure-related reticulo-thalamo-cortical
(RTC) malfunction. This issue also applies to the present study, and a comparative
investigation of the various subtypes within the GGE spectrum may clarify the underlying
cognitive pathophysiological mechanisms. Notably, a greater vulnerability of the auditory
information processing system to the effects of GGEs has been emphasized,[20] a fact linked to the administration modality of episodic memory tests.
Our findings regarding DM in adults with GGEs are in line with those of previous studies
conducted in pediatric populations,[16]
[41] which have consistently demonstrated deficits in semantic memory, reduced performance
in episodic memory, particularly in immediate and delayed episodic (story) recall,
and attention deficits. Notably, the latter may partially account for GGEs' reduced
episodic recall, as attention's pivotal role in memory performance, especially in
encoding, has been emphasized.[42] Moreover, seizure-induced neuronal dysfunction may exacerbate semantic deficits
by disrupting cognitive developmental trajectories,[8] therefore affecting semantic memory as well. Moreover, studies into the effects
of ASMs suggest that chronic treatment may detrimentally influence cognitive development.[43]
The ASEM deficit observed in our GGE patients aligns with previous and recent neuropsychological
findings in adult JME patients, suggesting impairment in associative semantic processing.[17]
[19]
The meta-analysis by Loughman et al.[44] provided evidence for consistent impairments in semantic knowledge and problem-solving
skills, which recapitulates findings in mixed GGE samples. The latter exhibit reduced
ability to manipulate acquired information, that is, semantic knowledge. Significantly
lower scores in GGE compared to HCs were also found on measures such as the Wechsler
Adult Intelligence Scale -WAIS Vocabulary and Information subtests, which gauge the
extent of semantic knowledge and factual DM respectively.[44]
Although DM has received considerable attention in the neuropsychological literature
on GGEs,[13] the autobiographical components have been overlooked. The present study comparing
GGE patients to HCs revealed impaired performance across two AEM and ASEM periods.
While delving into cognitive theories of memory exceeds the scope of the current study,
it is noteworthy to mention that we did not observe the anticipated ASEM advantage
in GGEs, as suggested by previous evidence for a double representation (double resistance)
of semantic information.[45] This could account, in the case of the present study, for the construct of ASEM
(knowledge about the self).[46] Moreover, since the GGE group exhibited impairments in the two AEM periods, the
temporal gradient hypothesis (that is, temporally-graded retrograde memory loss with
a disproportionate impairment of memory for events from the recent past relative to
remote memories) did not find support.
Compared to ASEM, AEM presents a more complex and demanding system, as it requires
control of the stored information, its temporal placing, processing, and confirmation
via the prefrontal cortex: individuals must recall the elements of an event – time,
place, and people – to mnemonically reconstruct the incident.[47] Memory recollections typically carry a significant emotional load, even when recalling
and describing a specific episode. Autobiographical episodic memory includes various
components, such as visual, verbal, emotional elements, whose activation involves
a complex process requiring participation from visual associative areas, the amygdala,
and frontal lobes.[48] In their AM model, Conway and Pleydell-Pearce[28] claim that AM disorders may arise either from diffuse cortical (occipital or frontotemporal)
damage or disconnection.[28] The aforementioned findings are not surprising, as evidence suggests that generalized
seizures may affect diverse regions, such as areas of the dominant temporal lobe and
the frontal and parietal association cortices.[49]
Positron-emission tomography (PET) studies on AM in healthy subjects put forward that
autobiographical retrieval relies on conceptual knowledge, with components having
representation in the bilateral temporal lobe.[50] The detrimental impact of generalized seizures on focal brain regions, coupled with
the bilateral temporal distribution of AM representations, may account for AM deficits
in GGE patients. Moreover, evidence suggests that in JAE, interictal limbic dysfunction
may interfere with AM. The specific neurocognitive derailments characterizing EGTCA
patients require further elucidation before advancing scientifically sound hypotheses
regarding AM impairment.
Interestingly, our correlational data ([Table 4]) suggest a selective involvement of visuoperceptual functions during the AEM and
ASEM childhood periods. Positive correlations emerged with measures of visuoperceptual
organization in AEM and ASEM regarding childhood, and with picture recognition exclusively
for the latter. Early evidence suggests that individuals who report more subjectively
vivid imagery on common questionnaires tend to recall past events and imagine future
events with a greater number of visual and sensory details.[51]
Table 4
Pearson correlations among AMI scores performance on and neuropsychological tests
among GGE patients
|
AEM-childhood
|
AEM-recent past
|
ASEM-childhood
|
ASEM-recent past
|
|
LM story IRe-WMS-I
|
−0.054
|
−0.165
|
0.176
|
−0.028
|
|
SVF
|
0.366
|
−0.046
|
0.335
|
0.065
|
|
SVF-clusters
|
−0.0016
|
−0.050
|
0.133
|
−0.042
|
|
PVF
|
0.121
|
0.128
|
−0.072
|
−0.338
|
|
PVF-clusters
|
0.008
|
−0.138
|
0.013
|
−0.165
|
|
RBMT story-I ImRe
|
0.133
|
0.184
|
0.064
|
0.223
|
|
RBMT pictures
|
0.194
|
−0.254
|
0.531*
|
−0.037
|
|
RBMT faces
|
0.216
|
−0.194
|
0.214
|
0.229
|
|
RBMT story-I Dre
|
0.187
|
−0.234
|
0.482*
|
0.062
|
|
RBMT name
|
0.076
|
−0.487*
|
0.076
|
0.473
|
|
RBMT personal items
|
−0.040
|
−0.082
|
0.089
|
0.394
|
|
RCFT-copy
|
0.555*
|
−0.188
|
0.462*
|
−0.224
|
|
RCFT-DRe
|
0.351
|
0.275
|
0.100
|
−0.047
|
|
VPT1 Visual
|
0.428*
|
0.141
|
0.194
|
−0.220
|
Abbreviations: AEM, autobiographical episodic memory; AMI, Autobiographical Memory
Interview; ASEM, autobiographical semantic episodic memory; Dre, delayed recall; GGE,
genetic generalized epilepsy; ImRe, immediate recall; LM story I Re-WMS-I, Logical
Memory story immediate recall-Weschsler Memory Scale-I; PVF, phonemic verbal fluency;
RBMT story-I, RBMT-faces recall story I; RCFT, Rey-Osterrieth Complex Figure Test;
SVF, semantic verbal fluency; VPT, Visual Patterns test.
Note: *Correlation is significant at the 0.05 level (2-tailed).
Case studies involving patients with visual imagery impairment may provide valuable
insights into potential shared brain networks supporting mental imagery and AM, despite
being often anecdotal and complicated by multiple comorbidities. For instance, Ogden[52] reported a patient suffering from cortical blindness due to occipital infarctions,
along with simultaneous visual imagery loss and autobiographical amnesia. Future studies
in GGE patients should address the potential link implied here between early-life
AM representations and visuospatial cognition.
Based on recent findings,[20] we herewith assume that the known effects of aberrant reticulo-thalamo-cortical
dynamics inducing secondary-systemic cognitive impairment in GGEs may, among other
already exposed factors, codetermine global AM deficits. The present study, to the
best of our knowledge, is the first to focus on AM in adult GGE patients. The limitations
to consider include the small size of the epilepsy subgroups, which may limit the
generalization of findings, and the use of only one AM test. Additionally, due to
the young age of the GGE sample, we excluded the early adulthood period from the study
to prevent overlap between AEM and ASEM responses for adulthood and the recent past,
which could represent a further limitation.
These findings warrant confirmation through future research on the cognitive pathophysiological
mechanisms underlying GGE-related long-term memory impairments.
In conclusion, the current study showed that GGE patients (JME, EGTCA and JAE), exhibit
global AM impairment. When compared to HCs, GGE patients displayed notable difficulties
in recalling both episodic and semantic autobiographical information from childhood
and the recent past. Additionally, GGE patients exhibited significantly lower performance
in immediate and delayed episodic recall, visuospatial working memory, visuoperceptual
organization, face recognition memory, and verbal executive functions. The challenges
observed in both semantic and episodic components of AM in GGE patients may stem from
secondary manifestations of primary cortical tone deregulation inherent in the pathophysiology
of GGE, as shown elsewhere.[20] Furthermore, the diminished performance of patients in the verbal executive functions
and visuospatial working memory domains may be suggestive of frontal lobe dysfunction,
which is particularly prominent in JME. Finally, the study highlights a specific impairment
in visuoperceptual functions related to retrieving autobiographical episodic and semantic
information from childhood, possibly implying a rapport between early-life AM systems
and visual cognition.
Bibliographical Record
Panayiotis Patrikelis, Eleni Loukopoulou, Elvira Masoura, Vasiliki Folia, Grigoris
Kiosseoglou, Lambros Messinis, Sonia Malefaki, Giuliana Lucci, Vasileios Kimiskidis.
Autobiographical memory impairment in genetic generalized epilepsies: neurocognitive
and pathophysiological determinants. Arq Neuropsiquiatr 2025; 83: s00451804923.
DOI: 10.1055/s-0045-1804923
