**4. P 300 abnormalities**

304 Neuroimaging for Clinicians – Combining Research and Practice

are dependent on the impairment of dopaminergic neural structures which regulate spatial frequency sensitivity. VEP findings were robustly confirmed in studies performed in animal models of PD [18, 19]. Despite the interest for the finding, only few confirmative studies were provided [15-17], most studies come from few laboratories devoted to this experimental approach. Guidelines for the use of VEPs in clinical practice only rarely suggested a role of VEPs in PD studies and VEPs were finally confined to the assessment of Multiple Sclerosis. Although the increased latency of the VEP in multiple sclerosis has in general been attributed to demyelination in the visual pathways, other mechanisms such as humoral factors, synaptic malfunction or changes in dendritic potentials may play a part. Such mechanisms may also be relevant to central nervous system disorders other than multiple sclerosis which have abnormal VEPs. Although the major clinical manifestations of PD involve the motor systems and the responsible pathology is located in the basal ganglia, there is evidence of more widespread disease, both pathologically, electrophysiologically and clinically. Only two studies explored the possible use of VEPs for the assessment of DLB [20,21] yet no comparison were presented with other forms of dementia. In MSA the visual system is believed to be spared and dopamine deficiency has been hypothesized to be less pronounced than in PD [22], even though the data in the literature are scarce and not unanimous and nothing on retinal dopamine content has been reported. Little information is available on VEPs and PERGs in MSA patients [22]. The main interest for studying responses elicited by patterns with pure chromatic contrast is that they allow recording of specific responses from colour-opponent pathways, anatomically and physiologically distinct from the achromatic ones at the retinal as well as the geniculate and cortical levels. A more recent study [23] showed that PERGs are virtually unaffected in MSA, whereas in early PD they are clearly impaired, suggesting different pathogenic retinal mechanisms and a useful simple tool for distinguishing MSA from PD. The strongest objection against the use of VEPs for the assessment of synucleinopathies derived from the technical constraints of VEP recordings: VEPs are altered by abnormalities of optic nerve and visual pathways, VEPs recordings require the adequate collaboration of patients who must focus attention on stimuli [24-26], VEP variable (amplitude, latency) are dependent on laboratories settings and

must be adjusted according to each laboratory statistics of distribution.

provide sufficient robustness in order to constitute a biomarker.

patient and control populations.

than improving [27].

The characteristic of VEP cannot be simply shared by different laboratories and differences in equipments might sustain variability which are far wider than variability observed in

With the introduction of digital and led stimulating screen this condition worsened rather

In the age range of AD, DLB and PDD optic and visual abnormalities (cataracts, maculopathies, retinopathies, ischemic lesions) are frequent and might mislead possible diagnoses. In DLB, fluctuations of cognition (i.e. defective attention and collaboration to the task) are common and might impair the diagnostic yield of VEP recordings. It has been suggested that the discrepancies between different reports on VEP in Parkinson's disease may be due to the greater sensitivity of grating patterns compared to checkerboard patterns and, if so, this might in part account for our normal PD VEP latencies. The grating subtense used is about one third of that for the checkerboard, and the retinal field stimulated is predominantly foveal for the grating whereas for the checkerboard it extends beyond the perimacula. This could explain the observed differences rather than the pattern form, per se. We suggest that VEP recordings might still represent a "niche" research tool, but do not In an event related potential (ERP) the P300 is a positive deflection peaking at approximately 300 ms after a stimulus. It is supposed to be an endogenous response, mainly depending on the processing of stimulus context, involving registration, evaluation and memory of stimuli, and categorization (decision/closure) and impinging on attention and arousal [28]. P300 can reliably be elicited with relatively simple paradigms, such as the "oddball paradigm", which requires the detection of a rare ("target") stimulus within a train of frequent irrelevant "non target" stimuli. Other complex paradigms include the administration of multiple stimuli, dichotic stimuli, multisensory modalities, in order to evaluate responses evoked from anterior brain regions. We recently performed a P300 study on patients with DLB in comparisons with patients with AD matched for dementia severity and age and with age matched control subjects [29] to look for differences of P300 responses in the two dementia subtypes and for possible correlations between P300 recordings and EEG, as abnormal EEG variability was described in DLB [5].

P300 responses were recorded with Ag/AgCl electrodes from 19 derivations corresponding to Fp1, Fp2, F3, F4, C3, C4, P3, P4, O1, O2, F7, F8, T3, T4, T5, T6, Cz, Fz, Pz positions of the 10-20 International System with supplementary A1 and A2 derivations. Care was taken to avoid recordings if variations of body temperature, recent food ingestion, or previous-night sleep disturbances were present [30]. A classical auditory "oddball" paradigm was used. The stimuli were 500 Hz and 1000 Hz tones, designated as the "non-target" and "target" stimuli, respectively, and delivered by STIM System Headphones. Patients were instructed to count only "target" stimuli, aloud in a preliminary trial and mentally in subsequent trials. The presentation ratio of "non target/target" tones for training was 4/1-8/1 and during recording purposes 5/1. The intensity of the tone was 75 dBnHL, the duration of the stimulus was 150 ms (rise-fall and the plateau times 5 and 140 ms, respectively). The presentation rate was random with a minimum inter-stimulus interval of 1.1 seconds and a maximum interval of 4 seconds. Digital filters was set at 0.15 Hz and 100 Hz, and averaged with a dwell time of 0.5 ms, 2000 Hz sampling rate, 100 ms of pre-stimulus baseline recording. An artifact rejection system was calibrated on four supplementary derivations placed on eyebrows and inferior orbital ridges; the rejection system blocked the acquisition when eye movement exceeded 100 uV. As the mean reference is known to distort P300 distribution one earlobe was used as online reference, with offline averaging with the other earlobe [31].

In each recording session, the correspondence between the counted and delivered stimuli was checked as described in previous studies [32] and stored on the hard disk. Sets where two or more targets had not been recognized or sets impaired by attention defects or false recognitions were discarded from analysis.

In each patient and control 120 responses to non-target and target stimuli were averaged in a single Final Average (FA). Four Sub-Averages (SA) of only 30 responses to target stimuli were preliminarily obtained in order to assess reliability of P300 among the recording sessions. Finally, inter-subject Grand Averages (GA) were obtained in DLB, AD and control groups. N100, N200, and P200 were detected at the Fz, Cz and Pz electrode for each subject separately. Peak latencies of each component were measured from stimulus onset to the point of maximum voltage in the range of 50-150 ms and 150-250 ms respectively. P300 was identified, in a time window of 300-500 ms, according to the operating definition based on

Is There a Place for Clinical Neurophysiology Assessments in Synucleinopathies? 307

[34], in the, supposedly, same kind of patients. The ''classic'' evaluation method was however encouraged by methodological guidelines [30] even after that topographic studies had been developed. The numerous clinical P300 studies suggest that this ERP component, elicited by auditory, visual, olfactory or somatosensory stimuli may be clinically useful as an index of such cognitive functions as attention and working memory. This assumption suggests that specific alterations should be found in DLB, where the cognitive disturbance is mainly characterized by fluctuating alterations of arousal and vigilance. Due to frontal dysexecutive dysfunction, DLB patients would be expected to express prevalent alterations of the anterior P3 component with a fronto-central scalp topography, whereas AD patients, with their early hypometabolism in the temporo-parietal junction, would be expected to show prevalent alterations in the parietal P3b response. If we restrict P300 measurements to classic assessment of P300 latencies in posterior (parietal) derivations, our study [32] shows that delayed latency and reduced amplitude, present in both dementia groups, can distinguish DLB from AD group, even though it is not possible to infer the applicability of these measures to an individual patient-to-patient analysis. The use of an active task did not allow us to investigate possible differences between groups in the mismatch negativity response, but we made sure that patients kept constant their attention during recordings, as P300 amplitude is sensitive to the amount of attention resources engaged during the task. In every group, N200 latencies were correlated with P300 latencies, confirming previous studies that showed prolonged N200 and P300 latencies in patients with dementia [46]. Topographical analysis of P300 recordings including all scalp leads did not add information about possible differences between patient groups, confirming that study of P300 topography could be limited to midline electrodes. Topographical differences, as latency distribution gradient, emerged and showed that P300 is different in DLB as compared to AD (figure 1): DLB patients had a more delayed P300 in anterior than in posterior derivations, while in all but two AD patients the latency was increased in posterior leads as compared to anterior leads, same as in controls. The normal latency distribution gradient consisting of increased latency in posterior leads as compared to anterior leads was reversed in DLB. Also the amplitude distribution gradients were reversed and thus different in DLB patients compared to AD or controls. The amplitude of P300 was prominent in frontal leads in DLB and in parietal leads in AD and controls. The finding of reversed amplitude gradient, with higher amplitude in frontal leads and smaller amplitude in posterior leads in DLB patients is apparently counterintuitive, as reduced amplitudes would be expected in a disease characterized by early frontal lobe involvement. Yet, delayed P300 latencies are also prominent in anterior leads of DLB patients and the two findings together seem compatible with the early frontal involvement of DLB. These findings suggest abnormal activity in anterior cortical areas of DLB patients, as compared with AD and controls. The correlation between P300 frontal delay and neuropsychological test scores exploring frontal lobe functions (FAB, NPI) supports this hypothesis. A possible interpretation might suggest that, in the early course of their disease, DLB patients need to increase efforts in frontal areas involved in recognition-attention tasks. P300 amplitude increment with delayed latency is

correlated to increments of encoding loads in experimental paradigms [47].

An alternative hypothesis could be that altered topographical P300 distribution in DLB represents a constant interference of the frontal P3a component, which is normally evoked by "novel" stimuli. According to this hypothesis DLB patients might produce frontal P300 component as the target stimuli will be interpreted as novel because DLB patients could not act and decide on these stimuli (i.e. match and encode in the target category). Further

its scalp distribution (central-parietal amplitude gradient), probability and sequence of preceding components. For every electrode location the following P300 variables were analyzed: amplitude (voltage difference between pre-stimulus baseline and the largest positive-going peak of the ERP waveform within a latency range of 300-500 ms), latency (time from the stimulus onset to the point of maximum positive amplitude within 300-500 ms time window: latencies were considered delayed if the peak latency was at least 2SD longer than the controls mean value. 2SD was chosen because previous studies showed that more restrictive criteria, 3SD from the mean, are insensitive to detect differences between controls and patients populations [33]), inter-electrode (Fz-Cz; Fz-Pz; Cz-Pz) latency and amplitude distribution gradients (difference in latency or amplitude of P300 responses between each pair of leads). The use of 3SD in defining normative limits for P300 was discharged as being at risk of excluding an excessive number of patients from investigative categorization. Studies on P300 scalp distribution and topography showed that peak latency changes across the scalp i.e., it is shorter over anterior cortical regions and longer over parietal areas [34], which allows to identify an earlier anterior, and a late posterior P3 component. Whether anterior and posterior P3 components recorded with a simple oddball paradigm originate from the same generators as those proposed for the P3a and P3b components obtained with the three-stimuli "novel" paradigm is yet unclear, however studies evaluating P3 scalp distribution suggested that early P3 component and late P3 component have separate origins: anterior superior temporal gyrus [35], prefrontal cortex [36], and anterior cingulate or supplementary motor area [37] for the early component vs temporo-parietal junction for the late one [38]. P300 studies in dementia were originally based on recordings from midline scalp derivations (the three leads Fz, Cz, Pz). Recordings in patients with dementia were compared with normative data obtained through P300 measurements in age-matched control populations. In demented patients, comparisons between ranges of different widths evidenced that, for P300 latency, the 2SD criterion had the greatest sensitivity in the detection of dementia [39]. Thus, if recordings in patients exceeded the 95% odds ratio, corresponding to normal mean±2SD, these recordings were considered abnormal and related to the cognitive disorder. By inference, delayed P300 latencies (by 2SD) or reduced P300 amplitudes were considered features of dementia, and thus useful diagnostic tools [40]. Based on this method, several studies found delays or amplitude decreases of P300 recorded from posterior (Pz) derivations in patients with putative AD [41], subcortical dementia [42], metabolic disorders [43] e PDD [44]. Yet, dementia categorization in the last ten years has been revolutionized by the identification of DLB, representing from 25 to 43% of all dementia cases. Therefore, one can assume that a discrete percentage of patients classified as AD patients in earlier studies, were instead affected by DLB. Because of cognitive ERPs were less investigated in DLB than other types of dementia, we examined the rates and qualitative features of P300 abnormalities in DLB vs AD patients. As EEG abnormalities are prevalent and linked to variability in DLB, the possible identification of correlations with EEG frequencies might support or challenge recent hypotheses suggesting that P300 is, or is not, the result of EEG phase resetting, due to orientation of attention to stimuli [45]. As our P300 recordings were obtained from a multielectrode montage covering the scalp, we could extend our analysis to further measurements, including topographic distribution of P300. Earlier topographical studies on P300 distribution were focused on AD patients, yet the same possible diagnostic flaws underlined above could be reported for topographic studies, and results were in some cases inconclusive with abnormal distributions described in anterior or in posterior derivations

its scalp distribution (central-parietal amplitude gradient), probability and sequence of preceding components. For every electrode location the following P300 variables were analyzed: amplitude (voltage difference between pre-stimulus baseline and the largest positive-going peak of the ERP waveform within a latency range of 300-500 ms), latency (time from the stimulus onset to the point of maximum positive amplitude within 300-500 ms time window: latencies were considered delayed if the peak latency was at least 2SD longer than the controls mean value. 2SD was chosen because previous studies showed that more restrictive criteria, 3SD from the mean, are insensitive to detect differences between controls and patients populations [33]), inter-electrode (Fz-Cz; Fz-Pz; Cz-Pz) latency and amplitude distribution gradients (difference in latency or amplitude of P300 responses between each pair of leads). The use of 3SD in defining normative limits for P300 was discharged as being at risk of excluding an excessive number of patients from investigative categorization. Studies on P300 scalp distribution and topography showed that peak latency changes across the scalp i.e., it is shorter over anterior cortical regions and longer over parietal areas [34], which allows to identify an earlier anterior, and a late posterior P3 component. Whether anterior and posterior P3 components recorded with a simple oddball paradigm originate from the same generators as those proposed for the P3a and P3b components obtained with the three-stimuli "novel" paradigm is yet unclear, however studies evaluating P3 scalp distribution suggested that early P3 component and late P3 component have separate origins: anterior superior temporal gyrus [35], prefrontal cortex [36], and anterior cingulate or supplementary motor area [37] for the early component vs temporo-parietal junction for the late one [38]. P300 studies in dementia were originally based on recordings from midline scalp derivations (the three leads Fz, Cz, Pz). Recordings in patients with dementia were compared with normative data obtained through P300 measurements in age-matched control populations. In demented patients, comparisons between ranges of different widths evidenced that, for P300 latency, the 2SD criterion had the greatest sensitivity in the detection of dementia [39]. Thus, if recordings in patients exceeded the 95% odds ratio, corresponding to normal mean±2SD, these recordings were considered abnormal and related to the cognitive disorder. By inference, delayed P300 latencies (by 2SD) or reduced P300 amplitudes were considered features of dementia, and thus useful diagnostic tools [40]. Based on this method, several studies found delays or amplitude decreases of P300 recorded from posterior (Pz) derivations in patients with putative AD [41], subcortical dementia [42], metabolic disorders [43] e PDD [44]. Yet, dementia categorization in the last ten years has been revolutionized by the identification of DLB, representing from 25 to 43% of all dementia cases. Therefore, one can assume that a discrete percentage of patients classified as AD patients in earlier studies, were instead affected by DLB. Because of cognitive ERPs were less investigated in DLB than other types of dementia, we examined the rates and qualitative features of P300 abnormalities in DLB vs AD patients. As EEG abnormalities are prevalent and linked to variability in DLB, the possible identification of correlations with EEG frequencies might support or challenge recent hypotheses suggesting that P300 is, or is not, the result of EEG phase resetting, due to orientation of attention to stimuli [45]. As our P300 recordings were obtained from a multielectrode montage covering the scalp, we could extend our analysis to further measurements, including topographic distribution of P300. Earlier topographical studies on P300 distribution were focused on AD patients, yet the same possible diagnostic flaws underlined above could be reported for topographic studies, and results were in some cases inconclusive with abnormal distributions described in anterior or in posterior derivations [34], in the, supposedly, same kind of patients. The ''classic'' evaluation method was however encouraged by methodological guidelines [30] even after that topographic studies had been developed. The numerous clinical P300 studies suggest that this ERP component, elicited by auditory, visual, olfactory or somatosensory stimuli may be clinically useful as an index of such cognitive functions as attention and working memory. This assumption suggests that specific alterations should be found in DLB, where the cognitive disturbance is mainly characterized by fluctuating alterations of arousal and vigilance. Due to frontal dysexecutive dysfunction, DLB patients would be expected to express prevalent alterations of the anterior P3 component with a fronto-central scalp topography, whereas AD patients, with their early hypometabolism in the temporo-parietal junction, would be expected to show prevalent alterations in the parietal P3b response. If we restrict P300 measurements to classic assessment of P300 latencies in posterior (parietal) derivations, our study [32] shows that delayed latency and reduced amplitude, present in both dementia groups, can distinguish DLB from AD group, even though it is not possible to infer the applicability of these measures to an individual patient-to-patient analysis. The use of an active task did not allow us to investigate possible differences between groups in the mismatch negativity response, but we made sure that patients kept constant their attention during recordings, as P300 amplitude is sensitive to the amount of attention resources engaged during the task. In every group, N200 latencies were correlated with P300 latencies, confirming previous studies that showed prolonged N200 and P300 latencies in patients with dementia [46]. Topographical analysis of P300 recordings including all scalp leads did not add information about possible differences between patient groups, confirming that study of P300 topography could be limited to midline electrodes. Topographical differences, as latency distribution gradient, emerged and showed that P300 is different in DLB as compared to AD (figure 1): DLB patients had a more delayed P300 in anterior than in posterior derivations, while in all but two AD patients the latency was increased in posterior leads as compared to anterior leads, same as in controls. The normal latency distribution gradient consisting of increased latency in posterior leads as compared to anterior leads was reversed in DLB. Also the amplitude distribution gradients were reversed and thus different in DLB patients compared to AD or controls. The amplitude of P300 was prominent in frontal leads in DLB and in parietal leads in AD and controls. The finding of reversed amplitude gradient, with higher amplitude in frontal leads and smaller amplitude in posterior leads in DLB patients is apparently counterintuitive, as reduced amplitudes would be expected in a disease characterized by early frontal lobe involvement. Yet, delayed P300 latencies are also prominent in anterior leads of DLB patients and the two findings together seem compatible with the early frontal involvement of DLB. These findings suggest abnormal activity in anterior cortical areas of DLB patients, as compared with AD and controls. The correlation between P300 frontal delay and neuropsychological test scores exploring frontal lobe functions (FAB, NPI) supports this hypothesis. A possible interpretation might suggest that, in the early course of their disease, DLB patients need to increase efforts in frontal areas involved in recognition-attention tasks. P300 amplitude increment with delayed latency is correlated to increments of encoding loads in experimental paradigms [47].

An alternative hypothesis could be that altered topographical P300 distribution in DLB represents a constant interference of the frontal P3a component, which is normally evoked by "novel" stimuli. According to this hypothesis DLB patients might produce frontal P300 component as the target stimuli will be interpreted as novel because DLB patients could not act and decide on these stimuli (i.e. match and encode in the target category). Further

Is There a Place for Clinical Neurophysiology Assessments in Synucleinopathies? 309

studies, in which infrequent distractor stimuli, will be inserted into the sequence of target and non target stimuli, should be carried out in order to evaluate the P3a and P3b components in AD and DLB patients. Indeed novel stimuli produce P3a component that is generally largest over the anterior and central recording sites and reflects frontal lobe function. On the other hand temporo-parietal pathway contributes to P300 from the target stimuli (P3b). Anyhow, the clinical utility of P300 recordings in differentiating DLB from AD was evidenced, in the patient populations with reliable P300 response, by sensitivity reaching 70% and specificity of 97%. Due to high specificity, when a reliable P300 is recorded in a patient with early dementia, and its gradients of latency and of amplitude across the scalp are reversed, i.e. anterior-to-posterior instead of the normal posterior to anterior distribution, P300 might have value to address diagnosis of DLB. Conversely, finding that P300 responses, although delayed and with reduced amplitude compared to controls, reach maximum amplitude and longer latencies in posterior leads suggests that the diagnosis of DLB is unlikely. The study of correlations between P300 recordings and neuropsychological test scores showed that increased latency and reduced amplitude were correlated with test scores assessing the presence of frontal lobe dysfunction (FAB), behavior abnormalities (NPI), fluctuating cognition (CAF). Topographical redistribution of P300 latency and amplitude, evidenced as distributions gradients were correlated with the presence of fluctuating cognition (positive CAF scores), typical symptom of DLB patients (figure 2). These correlations evidenced that the differences between groups are related to dementia and not to neuropsychiatric differences. A correlation between the performance of frontal lobe function in standardized neuropsychological tests and maximal P300 scalp distributions were also found in a previous study on a group of old adults [48]. Specifically, subjects who showed frontal-maximal P3 had lower performance than those elderly subjects who showed posterior-maximal scalp topographies. P300 measurements were also correlated with EEG descriptors (figure 2): latency and amplitude anterior to posterior distribution gradients were correlated with the DFP pre-alpha and with abnormal CSA patterns (CSA Patterns 2 to 4, see next on the test), typical of DLB, confirming the specificity

Patients with PD exhibit a reduced frequency of blinking leading to a staring appearance [49]. Reduced blink rate can cause an abnormal tear film, dry eyes and reduced vision. A characteristic ocular sign may be the blink reflex, elicited by a light tap on the glabella above the bridge of the nose: successive taps in normal individuals produce less and less response as the reflex habituates but in PD subjects the blink reflex does not disappear on repeated tapping. Habituation may improve after treatment with L-dopa or amantadine. Blink duration and excitability appear to be increased in PD and as in VEP latency may reflect loss of dopamine neurons [50]. The electric Blink Reflex (BR) is a neurophysiological technique exploring pontine structures through a reflex arc connecting nuclei of the 5th to the nuclei of the 7th cranial nerve. The Blink reflex consists of three separate responses: R1, R2, R3. The first one is generated in the trigemino-facial reflex arc, the second and third one are generated in polysynaptic pathways involving the brainstem reticular formation [51]. Clinically, the BR is used to evaluate brainstem lesions and it has been applied in clinical and neurophysiological studies of brainstem lesions and neurodegenerative disorders [52-54].

of topographical redistribution of P300 in DLB patients.

**5. Blink reflex abnormalities** 

Fig. 1. Grand averages and amplitude maps of P300 response in the three groups of subjects. A. Left. Grand averages of P300 responses in the DLB group. Vertical lines mark peak latency. U-shaped bars mark the difference in latency between Fz and Pz leads, same or shorter latency in Pz. Right. Amplitude map of P300 distribution throughout the scalp (at the maximum amplitude recorded) in DLB group. Notice anterior-to-posterior (reversed) amplitude distribution gradient. B. Left. Grand Averages of P300 responses in AD group. Vertical lines mark peak latency. U-shaped bars mark the difference in latency between Fz and Pz leads, longer in Pz. Right. Amplitude map of P300 distribution throughout the scalp (at the maximum amplitude recorded) in the AD group. Notice a posterior to anterior amplitude distribution gradient. C. Traces and distribution in controls. EOG: electrooculogram; DLB: Dementia with Lewy Bodies; AD: Alzheimer's Disease.

studies, in which infrequent distractor stimuli, will be inserted into the sequence of target and non target stimuli, should be carried out in order to evaluate the P3a and P3b components in AD and DLB patients. Indeed novel stimuli produce P3a component that is generally largest over the anterior and central recording sites and reflects frontal lobe function. On the other hand temporo-parietal pathway contributes to P300 from the target stimuli (P3b). Anyhow, the clinical utility of P300 recordings in differentiating DLB from AD was evidenced, in the patient populations with reliable P300 response, by sensitivity reaching 70% and specificity of 97%. Due to high specificity, when a reliable P300 is recorded in a patient with early dementia, and its gradients of latency and of amplitude across the scalp are reversed, i.e. anterior-to-posterior instead of the normal posterior to anterior distribution, P300 might have value to address diagnosis of DLB. Conversely, finding that P300 responses, although delayed and with reduced amplitude compared to controls, reach maximum amplitude and longer latencies in posterior leads suggests that the diagnosis of DLB is unlikely. The study of correlations between P300 recordings and neuropsychological test scores showed that increased latency and reduced amplitude were correlated with test scores assessing the presence of frontal lobe dysfunction (FAB), behavior abnormalities (NPI), fluctuating cognition (CAF). Topographical redistribution of P300 latency and amplitude, evidenced as distributions gradients were correlated with the presence of fluctuating cognition (positive CAF scores), typical symptom of DLB patients (figure 2). These correlations evidenced that the differences between groups are related to dementia and not to neuropsychiatric differences. A correlation between the performance of frontal lobe function in standardized neuropsychological tests and maximal P300 scalp distributions were also found in a previous study on a group of old adults [48]. Specifically, subjects who showed frontal-maximal P3 had lower performance than those elderly subjects who showed posterior-maximal scalp topographies. P300 measurements were also correlated with EEG descriptors (figure 2): latency and amplitude anterior to posterior distribution gradients were correlated with the DFP pre-alpha and with abnormal CSA patterns (CSA Patterns 2 to 4, see next on the test), typical of DLB, confirming the specificity of topographical redistribution of P300 in DLB patients.
