Current Alzheimer Research (www.bentham.org/car), 2004, 1, 63-69
Bentham Science Publishers Ltd.(www.bentham.org)

Cognitive Event-Related Potentials: Useful Clinical Information in Alzheimer’s Disease

Eiichi Katada1,*, Koichi Sato1, Kosei Ojika2 and Ryuzo Ueda3

1Department of Internal Medicine, Nagoya City Johoku Hospital, Japan, 2Department of Neurology and Neuroscience, Nagoya City University Graduate School of Medical Sciences, Japan, 3Department of Internal Medicine and Molecular Science, Nagoya City University Graduate School of Medical Sciences, Nagoya, Japan

*Address correspondence to this author at the Department of Internal Medicine, Nagoya City Johoku Hospital, 2-15, Kaneda-cho, Kita-ku, Nagoya 462-0033, Japan; Tel: +81-52-991-8121; Fax: +81-52-916-2038; E-mail: katada-ngi@umin.u-tokyo.ac.jp

Abstract: Cognitive event-related potentials (ERPs) have been used as a marker of cognitive function in patients with psychiatric and neurological disorders. In particular, the P300 potential has been widely utilized to study dementia and aging, because the P300 ERP component is easily observed and reflects attention and memory processing. However, the relationship between parameters of the P300 potential and the severity or type of cognitive impairment in patients with Alzheimer’s disease (AD) remains controversial. Because specific and effective anti-dementia treatments have recently become available for AD, more useful information is needed about the clinical aspects of this disease, including methods of making an accurate and early diagnosis, differentiation from vascular dementia and other degenerative dementias, assessment of severity, assessment of disease progression, evaluation of the response to treatment, and prediction of the prognosis. This mini-review described new discoveries on recent clinical application of ERPs in AD with respect to the above-mentioned areas. Although the definition of normal ERP values and the most appropriate method of ERP recording in AD patients have not been well defined, recent findings suggest that ERP analysis may be a clinically useful, inexpensive, noninvasive, and reliable method of assessing AD.

Keywords: Cognitive event-related potentials, P300, Alzheimer’s disease.

Introduction

Alzheimer’s disease (AD), one of the most common causes of mental deterioration in the elderly, is a progressive neurodegenerative disorder that is characterized by cognitive and behavioral dysfunction. After centrally acting cholinesterase inhibitors (ChEIs) became available, many recent studies have demonstrated their efficacy in AD patients. With regard to the early diagnosis of the detection of preclinical AD, mild cognitive impairment (MCI) has become important because it increases the risk of progression to overt AD [1]. Accordingly, reliable methods for the assessment of cognitive function in AD are required to make an accurate early diagnosis of AD, for differential diagnosis from vascular dementia and other degenerative dementias, for assessment of the severity, rate of progression, prognosis of AD, and for evaluation of the response to drug therapy. Various neuropsychological tests, including the Mini-Mental State Examination (MMSE), the Alzheimer’s Disease Assessment Scale-cognitive subscale (ADAS-cog), and Wechsler Adult Intelligence Scale (WAIS), have been used as indexes of cognitive function. However, these tests are subjective, and are often difficult to interpret. Cognitive event-related potentials (ERPs) have been used in psychiatric and neurological disorders as a marker of various cognitive functions. In particular, the P300 potential has been widely utilized to study dementia and aging, because the P300 ERP component is easy to observe and reflects attention and memory processing. Furthermore, the P300 has shown validity as an objective tool for the demonstration of cognitive function in AD (Fig. 1).

Fig. (1). Properties of event-related potentials (ERP).

Because the P300 has the longest history of clinical application among the cognitive ERPs, it is important to note that considerable progress in its use has been made, as has happened with other ERPs. It has generally been held that an ERP is the result of a set of discrete stimulus-evoked brain events.

Table 1.     Classification and Characteristics of Event-related Potentials (ERPs)

The ERPs provide evidence of a direct link between cognitive events and electrical activity in a wide range of cognitive paradigms (Tables 1 and 2).

Table 2.     The Identical Stimulus , Task, and Waveforms of Event-related Potentials (ERPs) by Each Paradigm

Therefore, ERPs are a useful supplement to neuropsychological tests that is the gold standard to assist clinical evaluations in AD patients. Although diagnostic use of ERPs must be guarded because of limited standardization and validation, information-processing analysis with ERPs may aid significantly in interpretation of behavioral data. This mini-review describes new discoveries in the field of cognitive ERPs in relation to AD.

Diagnostic utility of the P300 and limits to its clinical application in AD

Previous studies have demonstrated that the P300 amplitude decreases and the peak latency increases along with increasing severity of cognitive dysfunction [2, 3]. Although the P300 latency is a sensitive parameter for detecting early AD [4, 5, 6], its diagnostic utility has not yet been clarified because the abnormality rate of P300 latency in AD and other dementias ranges from 13% to 80% in different reports [4], probably because of differences in the method of recording P300 or in the clinical characteristics of A. The ERP resulting from averaging of repeated stimulus presentations varies in shape and amplitude in different recording loci, which gives a topographic indication of the localization of underlying brain activity. The data are from an auditory oddball paradigm and show the ERPs for the target stimulus.

B. The electrode at which maximum deflections are seen for a certain part of the waveform (i.e., Pz for P3) is selected to study the temporal sequence of processing events. Waveforms are numbered consecutively and prefixed by N (negative) or P (positive). The zero-line is defined by the voltage level of a reference electrode_which is often placed on the earlobes.the patients investigated in the various studies. Recently, to minimize the influence of external factors on the P300 and to maximize the ability to distinguish AD patients from normal persons, the procedures for recording P300 components and other ERPs have been improved by changing the stimulus characteristics, task factors, subject variables, method of electrophysiological recording, and method of analysis.

Specificity and sensitivity of ERP tests in AD

There is a growing consensus that validated molecular and biochemical markers of AD will complement clinical approaches by making early and accurate diagnoses. The ideal marker should have sensitivity greater than 80% for detecting AD and specificity also greater than 80% for distinguishing other dementias. The measurement of Ab42 and hyperphosphorylated tau protein in the cerebrospinal fluid (CSF) enables a fair distinction between AD and other neurodegenerative disorders [7, 8], reaching specificity and sensitivity levels of 85% [9]. Furthermore, the diagnostic accuracy of biomarker measurements may be improved by determination of the individual genetics background such as the apolipoprotein E genotype [10]. As for ERP tests, it is likely that the neuropathology of AD begins at the molecular/biochemical level where it interferes with synaptic mechanisms and may thus be first detected electro-physiologically (e.g., as in altered auditory ERPs). A more realistic approach to using ERPs in the diagnosis of dementia would be to base procedures on tests of cognitive function from experimental or clinical psychology. The oddball P300 has been considered as an indicator of memory function, however, the nature and specificity of the oddball P300 component has been controversial [11]. The design of ERP tests is a major problem because it is only possible to establish abnormality of the test in patients with diagnosed dementia, and so it is not known whether these tests are sensitive enough to identify someone with subclinical AD. The marker should be reliable, reproducible, non-invasive, simple to perform, and inexpensive. Thus, it is possible that the ERP could happen as in a biomarker of combing Ab and tau in CSF.

Recent advances in use of the P300 and other ERPs in AD

In general, auditory stimuli have been used because production is simple, and it is easy to obtain the attention of the subject. Many studies have investigated the utility of the auditory ERP for assessment of AD, but the relative sensitivity and specificity of the auditory ERP for detecting AD has not been shown to be high [12]. P300 waves triggered by novel stimuli show marked reduction in early AD relative to matched controls [13, 14]. Some studies have also investigated the optimal stimulus characteristics for visual, somatosensory, and olfactory modalities. Benvenuto et al. found that the statistical k nearest-neighbor method, which is one of the new methods of ERP recording that analyzes the potentials evoked by brief flashes of light, could distinguish AD patients from controls with a maximum sensitivity of 29% and a false-positive rate of 12%, while the comparable sensitivity/false-positive values for the statistical projection pursuit method and extended projection pursuit method (which selectively identify discriminative features) were 75%/18% and 100%/6%, respectively, indicating that a combination of selected ERP time segments from different electrodes contains, features that discriminate AD patients from normal subjects with a high sensitivity and specificity [15]. Coburn et al. reported that a potentially pathognomotic AD-specific signal was found within the flash visual-evoked potential (VEP), because there was selective delay of the P2 flash VEP component in AD patients compared with controls [16]. Although there was a significant between-group difference, however, the accuracy in individual patients and controls was too low for P2 delay to contribute meaningfully to the clinical diagnosis of AD. The olfactory ERP is one measure of olfactory function that demonstrates subtle changes of the olfactory system associated with normal aging [17]. As the processing of olfactory information occurs in areas of the brain that are affected in the early stage of AD, the olfactory ERP has been used to assess AD in recent years. Morgan et al. showed that the olfactory ERP P2 and P3 latencies were significantly longer in AD patients than in normal controls, that olfactory ERP latency was significantly correlated with the severity of dementia as measured by the Dementia Rating Scale, that olfactory ERP latency was superior at differentiating AD patients from normal controls than auditory ERP latency, that use of the olfactory ERP alone could correctly classify up to 92% of participants, and that combining the odor identification score with olfactory P3 latency resulted in a correct classification rate of 100% [18]. These findings indicate that the olfactory ERP is more useful and more sensitive than the commonly used auditory ERP for detecting subtle changes of the brain associated with AD. In early AD, impairment of episodic memory and semantic memory (knowledge normally retained in the long-term memory stores) occurs along with rapid forgetting of new memories, so physiological measures of episodic or semantic memory may prove useful in the early detection of AD. Rugg et al. suggested the late positive component (P600) could be a useful index of verbal episodic memory processes [19]. On the other hand, the N400 component is sensitive to semantic manipulations. Previous studies have shown that the N400 amplitude is decreased for words that are semantically in the correct context, but increases when words occur out of context. Beuzeron-Mangina et al. suggested that the P400 amplitude was significantly higher at posterior recording sites in patients with early AD than in age-matched normal subjects using an original memory workload paradigm, providing the first evidence that ERPs to their novel paradigm could be used as sensitive neurophysiological diagnostic markers of the brain changes underlying early AD as opposed to the memory processes of age-matched normal subjects [20]. Ford et al. recorded the ERPs to pictures and word targets (picture-name verification tasks) in patients with AD as well as in elderly and young controls, and demonstrated that N400 was larger (more negative) when the words did not match the pictures than when the words matched in all groups: In the young subjects, this difference was significant at all recordings sites, while this was only so at the central-parietal sites in the elderly and was limited to central-parietal sites on the right in AD patients [21]. Among AD patients pretested with a confrontation-naming task that involved identifying pictures they could not name, neither the N400 priming effect nor the scalp distribution of this potential was affected by the ability to name pictures correctly. This ERP evidence of spared knowledge of these items was associated with performance accuracy of 80%. Thus, although the name of an item may be inaccessible during confrontation naming, the N400 response show that enough memory remains to prime cortical responses. Moreover, Olichney et al. demonstrated that the latency of the N400 was slower in MCI, indicating that the congruous word repetition ERP effect appears to be sensitive to memory impairment in MCI and could be useful for predicting subclinical AD [22]. Moreover, Golob et al. demonstrated that there were significant differences of the increment in amplitude and latency of the P50 (sometimes called P1) component in MCI subjects during target detection compared with elderly controls, and that P300 latency was significantly longer in MCI, indicating that the brain potentials of MCI subjects during target detection have certain features similar to those in healthy elderly persons (RP, N100, P200, and N200), and other features similar to those in AD patients (delayed P300 latency, slower reaction time, and P50 differences) [23]. These differences might reflect pathophysiological changes in modulation of the auditory cortex by association regions that undergo neuropathological changes in early AD. By comparing the ERP changes of individuals with a positive family history of AD who carry the apolipoprotein E (APOE)e4 allele (E4+) and are at increased risk of developing AD with those of a group having a negative family history and a group with a positive family history but no APOEe4 allele, Green et al. showed that the P3 latency of the first group was significantly longer than that of the negative family history group at electrodes Pz and Cz. The N2 latency of the first subgroup was significantly longer than that of the second or third groups, indicating that individuals with an increased risk of AD show ERP changes consistent with those observed in patients diagnosed as having AD and that this disease has a preclinical phase when early detection may be possible [24]. Moreover, Tarkka et al. demonstrated that patients with familial AD showed a significantly smaller N100 peak amplitude and shorter latency of this automatic auditory evoked potential throughout the habituation test, when they compared three age-matched elderly groups, elderly patients with familial AD, elderly patients with sporadic AD, and healthy elderly subjects [25]. Such may be an objective indicator of impaired involuntary adaptation of the neuronal networks involved in auditory processing in this subtype of AD, and suggest that altered habituation may predict more rapid deterioration in familial AD.

Differential diagnosis of AD and other dementias

It is generally agreed that P300 parameters can differentiate demented subjects from healthy elderly persons, as well as from patients with alcoholism, depression, schizophrenia, Parkinson’s disease, supranuclear palsy, HIV dementia, and head injury [26]. However, the usefulness of P300 for the diagnosis of early AD is still controversial. Recently, Golob et al. detected an abnormal auditory P300 in subjects with MCI [23], while Frodl et al. demonstrated the early clinical diagnosis of AD using a combination of two auditory P300 subcomponents, i.e., the amplitude of the tempo-basal dipoles was significantly diminished in AD compared to age-matched healthy controls or MCI patients and the latencies of the tempo-superior dipoles were significantly prolonged in AD compared with age-matched healthy controls [27]. They stated that the sensitivity was 90.0% and the specificity was 79.1% for differentiation of AD patients from age-matched healthy controls. Thus, use of various stimuli processed in different ways might increase the diagnostic sensitivity of ERPs. With respect to the P3 component, two distinct subtypes are recorded with all stimulus modalities. One is a parietal maximal P300 (target P3), which is the response to task-relevant stimuli, and the other is a fronto-central P300 (novelty P3), which is considered to be a central nervous system index of the orienting response. Knight demonstrated that these two P3 components were useful for investigating the neural networks involved in specific cognitive processes [28]. Yamaguchi et al. suggested that the novelty P3 component, which was the response to novel task-irrelevant stimuli in an auditory oddball paradigm in contrast to the target P3, could discriminate AD from vascular dementia (VD), since the novelty P3 amplitude was preserved in AD patients and controls, while it was markedly reduced in VD patients, and the target P3 latency was prolonged in both AD and VD, whereas novelty P3 latency was only prolonged in VD [29]. However, it is not clear whether preservation of the responses to novel stimuli depends on the severity of dementia in AD patients. They also suggested that the response to novel stimuli might be relatively resistant to AD-related brain changes, at least in the early stage of the disease, because there was no correlation between the intelligence score and the novelty P3 latency or amplitude, so discordant P3 responses to target and novel stimuli in demented subjects might suggest a diagnosis of AD rather than VD.

P300 for evaluating the response of AD to drug therapy

Meador et al. [30] and Polich et al. [31] reported that the anticholinergic scopolamine delays the latency and decreases the amplitude of P300, while physostigmine (a cholinesterase inhibitor) reduces P300 latency over the short term, indicating that cholinergic neurons have an important role in the neuronal networks generating P300 potentials. Nunez showed that cognitive ERPs are largely comprised of summed excitatory and inhibitory postsynaptic potentials [32]. Concerning the neurotransmitters, there are GABAergic, cholinergic, noradrenergic, dopaminergic, and serotoninergic influences on P300 [33]. Polich et al. suggested that P300 could be an effective parameter for evaluating central nervous system medications, and that it would become much more important as the neurotransmitters involved in P300 generation became clearer [26]. The etiology of AD has not yet been elucidated, but the cholinergic hypothesis has been advanced on the basis of the presynaptic deficits found in the brain of AD patients and studies on the role of acetylcholine in animals and human behavior. Previous studies have demonstrated that centrally acting ChEIs can improve cognitive function in AD. However, evaluation of treatment has been mainly based on subjective assessment methods such as standardized neuropsychological tests including the MMSE, WAIS, and ADAS-cog. Therefore, it is necessary to obtain additional objective tools for evaluating the response of AD to drug therapy. Several reports have suggested that the P300 latency might provide useful information on the progression of AD, especially during the longitudinal follow-up of AD patients receiving treatment with ChEIs that act on the cholinergic pathways [34-39]. However, there are several problems related to P300 recording in patients with AD. A reliable P300 response to odd stimuli can not be recorded in some subjects and P300 is undetectable in a minority (5%-10%) of normal individuals. Moreover, neurophysiological changes do not necessarily correspond with clinical or behavioral changes. The results of neuropsychological tests sometimes remain abnormal after patients improve, while clinical improvement may not be significant despite improvement of the P300 latency. In general, P300 latency is considered a measure of stimulus classification speed and is generally unrelated to the response selection processes. Indeed, it is just these properties that make the P300 a valuable tool for assessing cognitive function: Because P300 latency is an index of processing time required before response generation, it is a sensitive temporal measure of neural activity underlying the processes of attention allocation and immediate memory. In addition, P300 latency is negatively correlated with mental function in normal subjects, with shorter latencies associated with superior cognitive performance. The neuropsychological tests that are best correlated with P300 latency are those that assess how rapidly subjects can allocate and maintain attentive resources. This association also supported by results indicating that P300 latency increases as cognitive capability decreases because of AD. Therefore, such discrepancies between the P300 latency and clinical findings may indicate that P300 latency reflects cognitive dysfunction more sensitively than clinical and behavioral features in AD patients. Otherwise, differences in findings by P300 from clinical and behavioral features suggest ERPs may be measuring a different domain whose clinical associations are still being characterized. Auditory sensory memory is one of the simplest types of short-term memory that can be studied electrophysiologically using mismatched negativity (MMN), which is a specific auditory ERP indexing automatic comparison of incoming stimuli to an existing memory trace. Previous studies have suggested that auditory sensory memory deteriorates with aging and especially in AD. However, it has not been decided whether MMN is regulated by the cholinergic system, the deterioration of which contributes to cognitive impairment in AD. Pekkonenet et al. recorded cortical auditory responses with a magnetometer after intravenous injection of various cholinergic antagonists, and suggested that the cholinergic system regulates the frequency-specific comparison of incoming stimuli with existing memory traces and modulates pre-attentive processing related to stimulus detection, and that the neural mechanisms responsible for cortical frequency- and duration-specific discrimination appear to have different sensitivities to cholinergic modulation [40]. Thus, their data obtained with a magnetometer supported the concept that auditory evoked potentials might be suitable for monitoring cholinergic activity in AD.

Prognostic electrophysiological mar-kers of AD

With regard to monitoring the progression of AD, there is evidence that the P300 latency increases as dementia becomes more severe and that the rate of increase reflects the rate of cognitive decline in both early-onset and late-onset AD [34-39,41,42]. Although abnormalities of the flash visual evoked portential (FVEP) latency are not consistently related to the severity of dementia [43], Swanwick et al. compared the annual rate of change of cognitive scales with baseline ERP/FVEP latencies and showed that patients with longer FVEP N2 component latencies had a slower subsequent rate of deterioration, suggesting that further study of FVEP responses as a prognostic marker of AD is warranted [44]. Saito et al. recorded visual ERPs and behavioral measures during a geometrical-figure discrimination task to examine sensory processing in patients with mild AD and age-matched controls, and found no differences between the groups with regard to the P1, N1, and P2 potentials, which reflect the early stage of sensory processing, as well as for NA potentials that reflect pattern recognition [45]. However, AD patients showed a reduced P3 amplitude, delayed reaction time, and increased behavioral errors compared with controls, indicating that patients with mild AD still had intact early sensory processing (including pattern recognition ), but showed selective compromise of higher-level processes (including integration of information and memory matching) that could influence behavioral deviation.

In conclusion, although the definition of normal ERP values and the most appropriate method of ERP acquisition in AD patients have not been clarified yet, recent reports suggest that the ERP may be a clinically useful, inexpensive, noninvasive, and reliable method for assessing AD.

Recent advances in our knowledge about P300 and other ERPs in AD patients as described in this mini-review are just part of a much larger research effort that promises to yield significant insights and future advances into the clinical application of ERPs.

References

[1]             Petersen RC, Smith GE, Waring SC, Ivnik RJ, Tangalos EG, Kokmen E. Mild cognitive impairment. Clinical characterization and outcome. Arch Neurol 56: 303-308 (1999).

[2]             Polich J, Ehlers CL, Otis S, Mandell AJ, Bloom FE: P300 latency reflects the degree of cognitive decline in dementing illness. Electroenceph Clin Neurophysiol 63: 138-144 (1986).

[3]             Ball SS, Marsh J, Schubarth G, Brown WS, Standburg R. Longitudinal P300 latency changes in Alzheimer’s disease. J Gerontol Med Sci 44: M195-200 (1989).

[4]             Polich J, Ladish C, Bloom F. P300 in early Alzheimer’s disease. Electroenceph Clin Neurophysiol 77: 179-189 (1990).

[5]             Holt L, Raine A, Pa G, Schneider L, Henderson V, Pollock VE. P300 topography in Alzheimer’s disease. Psychophysiology 32: 257-265 (1995).

[6]             Kazmerski VA, Friedman D. The scalp topography of P3b in early Alzheimer’s disease. J Psychophysiology 12: 127-143 (1998).

[7]             Hulstaert F, Blennow K, Ivanoiv A, Schoonderwaldt HC, Ricmenschneider M, De Deyn PP, et al. Improved discrimination of AD patients using beta-amyloid (1-42) and tau level in CSF. Neurology 52: 1555-1562 (1999).

[8]             Itoh N, Arai H, Urakami K, Ishiguro K, Ohno H, Hampel H, et al. Large-scale, multi-center study of cerebrospinal fluid tau protein phosphorylated at serine 199 for the antemortem diagnosis of Alzheimer’s disease. Ann Neurol 50: 150-156 (2001)

[9]             Andreas P, Christoph H. Biochemical markers of Alzheimer’s disease: wish and reality. Neurobiol Aging 23: 513-514 (2002).

[10]          Tapiola T, Piruila T, Mehta PD, Alafuzoff I, Lehtovirta M, Soininen H. Relationship between apoE genotype and CSF beta-amyloid (1-42) and tau in patients with probable and define Alzheimer’s disease. Neurobiol Aging 21: 735-740 (2000).

[11]          Verleger R. Event-related potentials and cognition: a critique of the context updating hypothesis and an alternative explanation of P3. Behav Brain Sci 11: 343-356 (1988).

[12]          Pokryszko-Dragon A, Slotwinski K, Podemski R. Modality-specific changes in P300 parameters in patients with dementia of the Alzheimer type. Med Sci Monit 9(4): 182-186 (2003).

[13]          Golob EJ, Starr A. Age-related qualitative differences in auditory cortical responses during short-term memory. Clin Neurophysiol 111: 2234-2244 (2000).

[14]          Daffiner KR, Rentz DM, Scinto LF, Faust R, Budson AE, Holcomb PJ. Pathophysiology underlying diminished attention to novel events in patients with early AD. Neurology 56: 1377-1383 (2001).

[15]          Benvenuto J, Jin Y, Casale M, Lynch G, Granger R. Identification of diagnostic evoked response potential segments in Alzheimer’s disease. Exp Neurol 176: 269-276 (2002).

[16]          Coburn KL, Arruda JE, Estes KM, Amoss RT. Diagnostic utility of visual evoked potential changes in Alzheimer’s disease. J Neuropsychiatry Clin Neurosci 15(2): 175-179 (2003).

[17]          Murphy C, Morgan CD, Geisler MW, Wetter S, Covington JW, Madowitz MD, et al. Olfactory event-related potentials and aging: Normative data. Int J Psychophysiol 36: 133-145 (2000).

[18]          Morgan CD, Murphy C. Olfactory event-related potentials in Alzheimer’s disease. J Int Neuropsychol Soc 8: 753-763 (2002).

[19]          Rugg MD, Roberts RC, Potter DD, Pickles CD, Nagy ME. Event-related potentials related to recognition memory: effects of unilateral temporal lobectomy and temporal lobe epilepsy. Brain 114: 2312-2332 (1991).

[20]          Beuzeron-Mangina JH, Mangina CA. Event-related brain potentials to Memory Workload and ‘Analytical-Specific Perception’ (Mangina-Test) in patients with early Alzheimer’s disease and in normal controls. Int J Psychophysiol 37: 55-69 (2000).

[21]          Ford JM, Askari N, Mathalon DH, Menon V, Gabrieli JD, Tinklenberg JR, et al. Event-related brain potentials evidence of spared knowledge in Alzheimer’s disease. Psychol Aging 16: 161-176 (2001).

[22]          Olichney JM, Morris SK, Ochoa C, Salmon DP, Thal LJ, Kutas M, et al. Abnormal verbal event related potentials in mild cognitive impairment and incipient Alzheimer’s disease. J Neurol Neurosurg Psychiatry 73: 377-384 (2002).

[23]          Golob EJ, Johnson JK, Starr A. Auditory event-related potentials during target detection are abnormal in mild cognitive impairment. Clin Neurophysiol 113: 151-161 (2001).

[24]          Green J, Levey AI. Event-related potential changes in groups at increased risk for Alzheimer’s disease. Arch Neurol 56: 1398-1403 (1999).

[25]          Tarkka IM, Lehtovirta M, Soininen H, Pääkkönen A, Karhu J, Partanen J. Auditory adaptation is differentially impaired in familial and sporadic Alzheimer’s disease. Biomed Pharmacother 56: 45-49 (2002).

[26]          Polich J, Herbst KL. P300 as a clinical assay: rationale, evaluation, and findings. Int J Psychophysiol 38: 3-19 (2000).

[27]          Frodl T, Hampel H, Juckel G, Bürger K, Padberg F, Engel RR, et al. Value of event-related P300 subcomponents in the clinical diagnosis of mild cognitive impairment and Alzheimer’s disease. Psychophysiology 39: 175-181 (2002).

[28]          Knight RT. Contribution of human hippocampal region to novelty detection. Nature 383: 256-259 (1996).

[29]          Yamaguchi S, Tsuchiya H, Yamagata S, Toyoda G, Kobayashi S, Event-related brain potentials in response to novel sounds in dementia. Clin Neurophysiol 111: 195-203 (2000).

[30]          Meador KJ, Loring DW, Davis HC, Sethi KD, Patel BRl. Cholinergic and serotonergic effects on the P3 potential and recent memory. J Clin Exp Neuropsychol 11: 252-260 (1989).

[31]          Polich J, Kok A. Cognitive and biologic determinants of P300: an integrative review. Bio Psychol 41: 103-146 (1995).

[32]          Nunez PL. Physical principles and neurophysiological mechanisms underlying event-related potentials. In: Rohrbaugh JW, Parasuraman R, Johnson R, eds. Event-related brain potentials. New York: Oxford University Press, 19-36 (1990).

[33]          Frodl-Bauch T, Bottlender R, Hegerl U. Neurochemical substrates and neuroanatomical generators of the event related P300. Neuropsychobiol 40(2): 86-94 (1999).

[34]          Reeves RR, Struve FA, Patricks G, Booker JG, Nave DW. The effects of donepezil on the P300 auditory and visual cognitive evoked potentials of patients with Alzheimer’s disease. Am J Geriatr Psychiatry 7: 349-352 (1999).

[35]          Thomas A, Iacono D, Bonanni L, D’ Andreamatteo G, Onofrj M. Donepezil, rivastigmine, and vitamin E in Alzheimer’s disease: a combined P300 event-related potential / neuropsychologic evaluation over 6 months. Clin Neuropharmacol 24: 31-42 (2001).

[36]          Onofrj M, Thomas A, Luciano AL, Iacono D, Di Rollo A, D’Andreamatteo G, et al. Donepezil versus vitamin E in Alzheimer’s disease: Part 2: mild versus moderate-severe Alzheimer’s disease. Clin Neuropharmacol 25(4): 207-215 (2002).

[37]          Weber AE, Klein C, Rabey JM. Evaluation of cholinergic treatment in demented patients by P300 evoked related potentials. Neur I Neurochir Pol Suplement 3: 37-43 (2001).

[38]          Weber AE, Gandelman-Marton R, Klein C, Rabey J. The clinical use of P300 event-related potentials for the evaluation of cholinesterase inhibitors treatment in demented patients. J Neural Transm 110: 659-669 (2003).

[39]          Katada E, Sato K, Sawaki A, Dohi Y, Ueda R, Ojika K. Long-term effects of donepezil on P300 auditory event-related potentials in patients with Alzheimer’s disease. J Geriatr Psychiatry Neurol 16(1): 39-43 (2003).

[40]          Pekkonen E, Hirvonen J, Jääskeläinen IP, Kaakkola S, Huttunen J. Auditory sensory memory and the cholinergic system: implications for Alzheimer’s disease. Neuroimage 14: 376-382 (2001).

[41]          St. Clair D, Blackburn I, Blackwood D, Tyrer G. Measuring the course of Alzheimer’s disease: a longitudinal study of neuropsychological function and changes in P3 event-related potential. Br J Psychiatry 152: 48-54 (1988).

[42]          Swanwick GR, Rowan MJ, Coen RF, Lawlor BA, Coakley D. Longitudinal event-related potentials in Alzheimer’s disease (letter). Br J Psychiatry 170: 580 (1997).

[43]          Swanwick GR, Rowan MJ, Coen RF, O’Mahony D, Lawlor BA. Longitudinal visual-evoked potentials in Alzheimer’s disease: a preliminary report. Biol Psychiatry 39: 455-457 (1996).

[44]          Swanwick GR, Rowan MJ, Coen RF, Coakley D, Lawlor BA. Prognostic value of electrophysiological markers in Alzheimer’s disease. Am J Geriatr Psychiatry 7: 335-338 (1999).

[45]          Saito H, Yamazaki H, Matsumoto K, Numachi Y, Yoshida S, Ueno T, et al. Visual event-related potential in mild dementia of the Alzheimer’s type. Psychiatry Clin Neurosci 55: 365-371 (2001).