Current Literature In Clinical Science
Who Am I? Forgetting Your Past Is Forgetting Yourself
Altered Self-Identity and Autobiographical Memory in Epilepsy. Allebone J, Rayner G, Siveges B, Wilson SJ. Epilepsia 2015;56:1982–1991
OBJECTIVE: Research suggests that individuals with chronic epilepsy display differences in their self-identity. The mechanisms by which self-identity is altered, however, are not well understood. Neural networks supporting autobiographical memory retrieval in the mesial temporal (MT) lobe are thought to be fundamental to self-identity processes. Thus, we examined differences in self-identity and autobiographical memory in patients with either MT or non–mesial temporal (NMT) foci with early or late age of habitual seizure onset. METHODS: Participants included 102 adults: 51 healthy individuals and 51 patients with drug-resistant focal seizures (19 MT, 32 NMT). We used the Ego Identity Process Questionnaire to profile the identity development of participants, and examined how this related to memory function assessed using the Autobiographical Memory Test. RESULTS: Patients and controls had strikingly different selfidentity profiles, with early onset MT patients showing the least identity development compared to controls and other patient groups. In contrast, late-onset NMT patients showed the highest level of identity development of the patient groups and closely resembled healthy controls (p < 0.05 for all comparisons). For all MT patients, poor autobiographical memory retrieval was correlated with altered self-identity (p < 0.001). No associations between autobiographical memory and self-identity were evident in the NMT group. SIGNIFICANCE: Self-identity in epilepsy may be modulated by the extent to which seizure foci impinge on the autobiographical memory network and the timing of seizure onset. Early disruption to MT regions of the autobiographical memory network may constitute a neurocognitive mechanism by which self-identity is altered in chronic focal epilepsy.
Commentary A sense of personal identity is intuitively a core component of being human, and autobiographical memory (AM), memory of personal experiences, is thought to be critical for identity development. It is unclear whether animals have such selfreferential memory. Barfield (1) noted that his cat became anxious when suitcases appeared, portending unwelcome solitude, but he could not conceive of his cat “sitting down and thinking to himself: ‘Let’s see, what was I doing this time yesterday?’” Allebone and colleagues have undertaken the task of relating AM to a sense of identity, and then to mesial temporal lobe epilepsy (MTLE). Every one of these concepts is difficult to measure. Nevertheless, this study is compelling because these factors are so important and so potentially relevant to the life history of persons with epilepsy. Most or our patients complain about their memory. There is some evidence that hippocampal damage affects AM more than semantic memory, the memory of impersonal facts (2). Allebone and colleagues hypothesize that AM impairment, as measured by the Autobiographical Memory Test (AMT), disrupts the complex process of self-identity formation, as measured by the Ego Identity Process Questionnaire (EIPQ). They Epilepsy Currents, Vol. 16, No. 3 (May/June) 2016 pp. 141–142 © American Epilepsy Society
suggest that MTLE damages AM, and this leads to the poor development of self-identity. Among their 19 subjects with MTLE, it was found that poor AM did indeed correlate with poor self-identity. Early-onset MTLE (before age 18), later onset MTLE, and epilepsy not localized to the mesial temporal region (non-MTLE) were all associated with impaired AM. However, only in the MTLE group was impaired AM strongly correlated with poor self-identity. This was especially true of early-onset MTLE. Why these two functions were linked in MTLE but not in other the groups, including the healthy controls, is a mystery. The terminology of memory can be confusing. Allebone et al. make a point of distinguishing between AM and semantic memory, but they also state that semantic memory is based on “episodic” memory. Of course, much of episodic memory is autobiographical. So AM and semantic memory must be somewhat intertwined. Of the two instruments used in this study, the AMT is easier to understand. Persons are given emotionally-valenced cue words (e.g., happy) and are asked to associate memories. The more specific the memory, such as of a single day, the higher the score (of course, the observer, and perhaps the subject, has no way of knowing whether the memory is true). Defining “identity” is even more challenging. What is meant by poor self-identity? It is important to understand that, as defined in this study, it does not necessarily mean either low self-esteem or a dissociative tendency. The EIPQ is an attempt to operationalize the nebulous concept of self-
141
Memory, Identity and Mesial Temporal Epilepsy
identity. Individuals are rated on two dimensions, commitment and curiosity (exploration). High commitment can be either “foreclosed,” presumably bad because it is premature, or “achieved,” presumably good. Low commitment can be either a state of “moratorium,” potentially good because the person is still exploring, or “diffused,” neither exploring nor committed. Adolescents, for example, would typically be in the “moratorium” category of low commitment but high exploration. Classification of a person into one of these four categories is based on responses involving the following four ideological domains: 1) occupation, 2) religion, 3) politics, and 4) values; and also four interpersonal domains: 1) family, 2) friendships, 3) relationships, and 4) gender roles. It should be apparent that several of these domains may be subject to value judgments. Americans assume that ego strength is a positive thing, but in many cultures, group identity trumps individual identity and may not be measurable in the same way. If we accept that a strong self-identity is a positive thing, then deciding which characteristics within these eight domains underpin it may be partly subjective. How does all this relate to epilepsy? The authors conclude that mesial temporal seizure foci disrupt the AM network during a critical time for formation of self-identity, such as adolescence, and that this disruption is causative for the identity dysfunction. This chain of events is plausible, but the facts tying MTLE to AM dysfunction and then to identity problems require confirmation. In this study, there were only 10 persons with early MTLE and 10 with late-onset MTLE. Laterality may be important (3), but there were only four left and four right early MTLE patients. The authors were careful not to equate MTLE with mesial temporal sclerosis (MTS), but the number who had MTS on a MRI was not specified. Finally, why is there no correlation between AM and self-identity in the non-MTLE group? Their AM was just as bad as the MTLE group. It is hard to make sense of this finding since the non-MTLE group included a grab bag of lateral TLE, frontal, parietal, and unknown seizure origins. Furthermore, the hypothesis that this association is critical during childhood should be studied in patients with non-mesial temporal early-onset epilepsy. The impairment of AM in all epilepsy groups calls into question the assumption that mesial temporal function is essential for personal recollection. Recent results of very selective laser ablation of mesial temporal structures suggest that they are not always critical for every memory function (3). The vital role of the temporal lobes in memory is an idea of surprisingly recent vintage. In ancient times, memory was
142
thought to be a function of the heart (thus, memorizing “by heart”), then later the cerebral ventricles. Shakespeare, following Galen, spoke of the “ventricle of memory” (4). In 1887, Brown and Schaefer (5) reported that a monkey, with both temporal lobes removed, acted as if “every object with which he comes in contact...appears strange and is investigated with curiosity.” However, it was not until Penfield’s time that human memory became closely associated with the temporal lobes. The tragic example of patient H.M., who lost essentially all autobiographical memory after surgical removal of both of his temporal lobes, did not occur until 1953 (6). It is now evident that memory is not a unitary function, nor are all aspects of memory associated with mesial temporal structures. We still have much to learn about the functional interconnections underlying various aspects of memory. The concept of selfidentity is central to many psychological theories but is hard to quantify. More research is needed to verify the linkage of various types of memory with one’s sense of self, to the implications for function and for quality of life, and to the degree to which these entities are affected by various epilepsies. The work by Allebone and colleagues supports the notion that some aspects of memory are necessary to our sense of who we are in the world. Impairment of that sense may be an underrecognized comorbidity, and a heavy burden, for persons with epilepsy. Perhaps we should modify Descartes’ aphorism to “Memorio Ergo Sum.” I remember, therefore I am. by Edward Faught, MD References 1. Barfield O. The ventricle of memory. Anthroposophical Quarterly 1975;20:10–11. 2. Addis DR, Moscovitch M, McAndrews MP. Consequences of hippocampal damage across the autobiographical memory network in left temporal lobe epilepsy. Brain 2007;130:2327–2342. 3. Drane DL, Loring DW, Voets NL, Price M, Ojemann JG, Willie JT, Saindane AM, Phatak V, Ivanisevic M, Mills S, Helmers SL, Miller JW, Meador KJ, Gross RE. Better object recognition and naming outcome with MRI-guided stereotactic laser amygdalohippocampectomy for temporal lobe epilepsy. Epilepsia 2015;56:101–113. 4. Shakespeare W. Love’s Labours Lost. Washington, D.C., Folger Digital Shakespeare Library; 2006: Act IV, Scene 2, 101, lines 85–86. 5. Finger S, Origins of Neuroscience. Oxford, Oxford University Press, 1994:365). 6. Scoville WB, Milner B. Loss of recent memory after bilateral hippocampal lesions. J Neurol Neurosurg Psych 1957;20:11–21.
Who Am I? Forgetting Your Past Is Forgetting Yourself
Altered Self-Identity and Autobiographical Memory in Epilepsy. Allebone J, Rayner G, Siveges B, Wilson SJ. Epilepsia 2015;56:1982–1991
OBJECTIVE: Research suggests that individuals with chronic epilepsy display differences in their self-identity. The mechanisms by which self-identity is altered, however, are not well understood. Neural networks supporting autobiographical memory retrieval in the mesial temporal (MT) lobe are thought to be fundamental to self-identity processes. Thus, we examined differences in self-identity and autobiographical memory in patients with either MT or non–mesial temporal (NMT) foci with early or late age of habitual seizure onset. METHODS: Participants included 102 adults: 51 healthy individuals and 51 patients with drug-resistant focal seizures (19 MT, 32 NMT). We used the Ego Identity Process Questionnaire to profile the identity development of participants, and examined how this related to memory function assessed using the Autobiographical Memory Test. RESULTS: Patients and controls had strikingly different selfidentity profiles, with early onset MT patients showing the least identity development compared to controls and other patient groups. In contrast, late-onset NMT patients showed the highest level of identity development of the patient groups and closely resembled healthy controls (p < 0.05 for all comparisons). For all MT patients, poor autobiographical memory retrieval was correlated with altered self-identity (p < 0.001). No associations between autobiographical memory and self-identity were evident in the NMT group. SIGNIFICANCE: Self-identity in epilepsy may be modulated by the extent to which seizure foci impinge on the autobiographical memory network and the timing of seizure onset. Early disruption to MT regions of the autobiographical memory network may constitute a neurocognitive mechanism by which self-identity is altered in chronic focal epilepsy.
Commentary A sense of personal identity is intuitively a core component of being human, and autobiographical memory (AM), memory of personal experiences, is thought to be critical for identity development. It is unclear whether animals have such selfreferential memory. Barfield (1) noted that his cat became anxious when suitcases appeared, portending unwelcome solitude, but he could not conceive of his cat “sitting down and thinking to himself: ‘Let’s see, what was I doing this time yesterday?’” Allebone and colleagues have undertaken the task of relating AM to a sense of identity, and then to mesial temporal lobe epilepsy (MTLE). Every one of these concepts is difficult to measure. Nevertheless, this study is compelling because these factors are so important and so potentially relevant to the life history of persons with epilepsy. Most or our patients complain about their memory. There is some evidence that hippocampal damage affects AM more than semantic memory, the memory of impersonal facts (2). Allebone and colleagues hypothesize that AM impairment, as measured by the Autobiographical Memory Test (AMT), disrupts the complex process of self-identity formation, as measured by the Ego Identity Process Questionnaire (EIPQ). They Epilepsy Currents, Vol. 16, No. 3 (May/June) 2016 pp. 141–142 © American Epilepsy Society
suggest that MTLE damages AM, and this leads to the poor development of self-identity. Among their 19 subjects with MTLE, it was found that poor AM did indeed correlate with poor self-identity. Early-onset MTLE (before age 18), later onset MTLE, and epilepsy not localized to the mesial temporal region (non-MTLE) were all associated with impaired AM. However, only in the MTLE group was impaired AM strongly correlated with poor self-identity. This was especially true of early-onset MTLE. Why these two functions were linked in MTLE but not in other the groups, including the healthy controls, is a mystery. The terminology of memory can be confusing. Allebone et al. make a point of distinguishing between AM and semantic memory, but they also state that semantic memory is based on “episodic” memory. Of course, much of episodic memory is autobiographical. So AM and semantic memory must be somewhat intertwined. Of the two instruments used in this study, the AMT is easier to understand. Persons are given emotionally-valenced cue words (e.g., happy) and are asked to associate memories. The more specific the memory, such as of a single day, the higher the score (of course, the observer, and perhaps the subject, has no way of knowing whether the memory is true). Defining “identity” is even more challenging. What is meant by poor self-identity? It is important to understand that, as defined in this study, it does not necessarily mean either low self-esteem or a dissociative tendency. The EIPQ is an attempt to operationalize the nebulous concept of self-
141
Memory, Identity and Mesial Temporal Epilepsy
identity. Individuals are rated on two dimensions, commitment and curiosity (exploration). High commitment can be either “foreclosed,” presumably bad because it is premature, or “achieved,” presumably good. Low commitment can be either a state of “moratorium,” potentially good because the person is still exploring, or “diffused,” neither exploring nor committed. Adolescents, for example, would typically be in the “moratorium” category of low commitment but high exploration. Classification of a person into one of these four categories is based on responses involving the following four ideological domains: 1) occupation, 2) religion, 3) politics, and 4) values; and also four interpersonal domains: 1) family, 2) friendships, 3) relationships, and 4) gender roles. It should be apparent that several of these domains may be subject to value judgments. Americans assume that ego strength is a positive thing, but in many cultures, group identity trumps individual identity and may not be measurable in the same way. If we accept that a strong self-identity is a positive thing, then deciding which characteristics within these eight domains underpin it may be partly subjective. How does all this relate to epilepsy? The authors conclude that mesial temporal seizure foci disrupt the AM network during a critical time for formation of self-identity, such as adolescence, and that this disruption is causative for the identity dysfunction. This chain of events is plausible, but the facts tying MTLE to AM dysfunction and then to identity problems require confirmation. In this study, there were only 10 persons with early MTLE and 10 with late-onset MTLE. Laterality may be important (3), but there were only four left and four right early MTLE patients. The authors were careful not to equate MTLE with mesial temporal sclerosis (MTS), but the number who had MTS on a MRI was not specified. Finally, why is there no correlation between AM and self-identity in the non-MTLE group? Their AM was just as bad as the MTLE group. It is hard to make sense of this finding since the non-MTLE group included a grab bag of lateral TLE, frontal, parietal, and unknown seizure origins. Furthermore, the hypothesis that this association is critical during childhood should be studied in patients with non-mesial temporal early-onset epilepsy. The impairment of AM in all epilepsy groups calls into question the assumption that mesial temporal function is essential for personal recollection. Recent results of very selective laser ablation of mesial temporal structures suggest that they are not always critical for every memory function (3). The vital role of the temporal lobes in memory is an idea of surprisingly recent vintage. In ancient times, memory was
142
thought to be a function of the heart (thus, memorizing “by heart”), then later the cerebral ventricles. Shakespeare, following Galen, spoke of the “ventricle of memory” (4). In 1887, Brown and Schaefer (5) reported that a monkey, with both temporal lobes removed, acted as if “every object with which he comes in contact...appears strange and is investigated with curiosity.” However, it was not until Penfield’s time that human memory became closely associated with the temporal lobes. The tragic example of patient H.M., who lost essentially all autobiographical memory after surgical removal of both of his temporal lobes, did not occur until 1953 (6). It is now evident that memory is not a unitary function, nor are all aspects of memory associated with mesial temporal structures. We still have much to learn about the functional interconnections underlying various aspects of memory. The concept of selfidentity is central to many psychological theories but is hard to quantify. More research is needed to verify the linkage of various types of memory with one’s sense of self, to the implications for function and for quality of life, and to the degree to which these entities are affected by various epilepsies. The work by Allebone and colleagues supports the notion that some aspects of memory are necessary to our sense of who we are in the world. Impairment of that sense may be an underrecognized comorbidity, and a heavy burden, for persons with epilepsy. Perhaps we should modify Descartes’ aphorism to “Memorio Ergo Sum.” I remember, therefore I am. by Edward Faught, MD References 1. Barfield O. The ventricle of memory. Anthroposophical Quarterly 1975;20:10–11. 2. Addis DR, Moscovitch M, McAndrews MP. Consequences of hippocampal damage across the autobiographical memory network in left temporal lobe epilepsy. Brain 2007;130:2327–2342. 3. Drane DL, Loring DW, Voets NL, Price M, Ojemann JG, Willie JT, Saindane AM, Phatak V, Ivanisevic M, Mills S, Helmers SL, Miller JW, Meador KJ, Gross RE. Better object recognition and naming outcome with MRI-guided stereotactic laser amygdalohippocampectomy for temporal lobe epilepsy. Epilepsia 2015;56:101–113. 4. Shakespeare W. Love’s Labours Lost. Washington, D.C., Folger Digital Shakespeare Library; 2006: Act IV, Scene 2, 101, lines 85–86. 5. Finger S, Origins of Neuroscience. Oxford, Oxford University Press, 1994:365). 6. Scoville WB, Milner B. Loss of recent memory after bilateral hippocampal lesions. J Neurol Neurosurg Psych 1957;20:11–21.
