- Email: [email protected]
Visuospatial information processing load and the ratio between parietal cue and target P3 amplitudes in the Attentional Network Test
Accepted Manuscript Title: Visuospatial information processing load and the ratio between parietal cue and target P3 amplitudes in the Attentional Network Test Author: Dimitri M. Abramov Monique Pontes Adailton T. Pontes Carlos A. Mourao-Junior Juliana Vieira Carla Quero Cunha Tiago Tamborino Paulo R. Galhanone Leonardo C. deAzevedo Vladimir V. Lazarev PII: DOI: Reference:
S0304-3940(17)30251-3 http://dx.doi.org/doi:10.1016/j.neulet.2017.03.031 NSL 32719
To appear in:
Neuroscience Letters
Received date: Revised date: Accepted date:
9-9-2016 2-2-2017 17-3-2017
Please cite this article as: D.M. Abramov, M. Pontes, A.T. Pontes, C.A. MouraoJunior, J. Vieira, C.Q. Cunha, T. Tamborino, P.R. Galhanone, L.C. deAzevedo, V.V. Lazarev, Visuospatial information processing load and the ratio between parietal cue and target P3 amplitudes in the Attentional Network Test, Neuroscience Letters (2017), http://dx.doi.org/10.1016/j.neulet.2017.03.031 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Ac
ce
pt
ed
M
an
us
cr
i
Graphical Abstract
Page 1 of 21
*Highlights
Highlights - The cue containing spatial information about the target increases parietal cue P3. - This spatial cue reduces the subsequent parietal target P3. - The magnitudes of cue and target P3 total the same in spatial and neutral cueing.
Ac
ce pt
ed
M
an
us
cr
- The parietal P3 reflects visuospatial information processing.
ip t
- The parietal cue P3 positively correlates with the ability to perform spatial tasks.
Page 2 of 21
*Manuscript Click here to download Manuscript: Manuscript revised i2017.doc
Click here to view linked References
Visuospatial information processing load and the ratio between parietal cue and target P3 amplitudes in the Attentional Network Test Dimitri M. Abramova, Monique Pontesa, Adailton T. Pontesa, Carlos A. Mourao-Juniorb, Juliana Vieiraa, Carla Quero Cunhaa, Tiago Tamborinoa, Paulo R. Galhanonea, Leonardo C.
a
ip t
deAzevedoa, Vladimir V. Lazareva,*. Laboratory of Neurobiology and Clinical Neurophysiology, National Institute of Women,
cr
Children and Adolescents Health Fernandes Figueira, Oswaldo Cruz Foundation
(FIOCRUZ), Av. Ruy Barbosa, 716, Flamengo, Rio de Janeiro, RJ, 22250-020, Brazil. Department of Physiology, Federal University of Juiz de Fora, Campus UFJF - ICB, Juiz de
us
b
Corresponding author. E-mail address: [email protected] (V.V. Lazarev).
ed
M
*
an
Fora, MG, 36036-900, Brazil.
Abstract
ce pt
In ERP studies of cognitive processes during attentional tasks, the cue signals containing information about the target can increase the amplitude of the parietal cue P3 in relation to the ‘neutral’ temporal cue, and reduce the subsequent target P3 when this
Ac
information is valid, i.e. corresponds to the target’s attributes. The present study compared the cue-to-target P3 ratios in neutral and visuospatial cueing, in order to estimate the contribution of valid visuospatial information from the cue to target stages of the task performance, in terms of cognitive load. The P3 characteristics were also correlated with the results of individuals’ performance of the visuospatial tasks, in order to estimate the relationship of the observed ERP with spatial reasoning. In 20 typically developing boys, aged 10–13 years (11.3 ± 0.86), the intelligence quotient (I.Q.) was estimated by the Block Design and Vocabulary subtests from the WISC-
Page 3 of 21
2 III. The subjects performed the Attentional Network Test (ANT) accompanied by EEG recording. The cued two-choice task had three equiprobable cue conditions: No cue, with no information about the target; Neutral (temporal) cue, with an asterisk in the center of the visual field, predicting the target onset; and Spatial cues, with an asterisk in the upper or lower
ip t
hemifield, predicting the onset and corresponding location of the target. The ERPs were estimated for the mid-frontal (Fz) and mid-parietal (Pz) scalp derivations.
cr
In the Pz, the Neutral cue P3 had a lower amplitude than the Spatial cue P3; whereas
us
for the target ERPs, the P3 of the Neutral cue condition was larger than the Spatial cue P3. However, the sums of the magnitudes of the cue and target P3 were equal in the spatial and
an
neutral cueing, probably indicating that in both cases the equivalent information processing load is included in either the cue or the target reaction, respectively. Meantime, in the Fz, the
M
analogue ERP components for both the cue and target stimuli did not depend on the cue condition. The results show that, in the parietal site, the spatial cue P3 reflects the processing
ed
of visuospatial information regarding the target position. This contributes to the subsequent "decision-making", thus reducing the information processing load on the target response,
ce pt
which is probably reflected in the lower P3. This finding is consistent with the positive correlation of parietal cue P3 with the individual's ability to perform spatial tasks as scored by
Ac
the Block Design subtest.
Key words: visuospatial information processing; event related potentials; P3; cueing of target; parietal cortex; Attentional Network Test.
Page 4 of 21
3 Introduction The positive component of the event-related potential (ERP), which appears 300-500 ms after the event occurrence, is known as P3 or P300 [1-3]. This wave is related to a subjective estimation of the signal. It is usually elicited in goal-directed behavior during
ip t
attentional tasks [2,3], such as the OddBall [4], Go-NoGo [5], and other paradigms. It shows high sensitivity to the properties of a task, such as novelty, difficulty, and so on, and correlate
cr
with characteristics of attention and control [2,3,6,7].
us
The P3 is associated with the activity of temporo-parietal networks connected with the frontal cortex and dependent on the modality of a sensory stimulus [3,8,9]. The maximum
an
manifestation of this component is recorded in the parietal region [3] above the associative temporo-parieto-occipital cortex (TPO), which is responsible for the spatial processing of the
M
visual system [10,11], integration of information for action, and simultaneous spatial/quasispatial types of syntheses [11-13]. In general, these cortical areas relate to encoding, storage
ed
and retrieval of information [6,12].
ce pt
The cues of different modalities containing information about the target evoke higher amplitudes of the cue P3 [14,15] than the "neutral" temporal cue, which predicts the moment of the target's appearance. The valid cue signals corresponding to the target attributes can lead to reduction of the target P3 amplitude [16], which apparently reflects a higher probability of
Ac
the target appearing (according to [1,17]), and less subjective efforts in perceptual processing, attention orientation and control, probably determined by information that the cue conveyed. Importantly, this inverse relationship between the cue and target P3 is apparent when compared with the neutral temporal cueing. This follows from several studies, which, however, did not assess this effect directly, describing the cue or target P3 separately in the framework of other research objectives [18] or without neutral temporal cueing as a reference [16,19]. However, a temporally cued target P3 does not always show a difference from a
Page 5 of 21
4 validly cued target P3 [14]. This problem deserves special investigation, since the valid-cueing paradigm allows one to track the information-processing load from the cue to the target stage of the task performance. The present study evaluated the ratio between the cue and the target P3, based on the
ip t
Posner`s Attentional Network Test (ANT) [18,20-22]. The ANT operates with neutral temporal, valid spatial and no-cueing of the target onset and/or location in the visual field. We
cr
used this approach to detect the putative manifestation of the transfer of valid visuospatial
us
information from cue to target ERPs, and to estimate the contribution of this transfer to the final response in terms of cognitive load [17].
an
These terms refer to the total amount of mental efforts or neural resources being used during performance of the task [23]. The P3 is considered as a sensitive workload index
M
reflecting demand for information processing resources [17]. A likely reallocation of a portion of the workload from one task to a simultaneously executed another was shown by means of
ed
the P3 amplitude changes dependent on the first task’s difficulty and probably caused by a subject’s fixed capacity of neural resource available for the given type of mental operations
ce pt
[17,24]. We expected that the cue-to-target dynamics of the P3 amplitude in the ANT paradigm also could be adequately described by the workload. The latter might be a general and comprehensive psychophysiological measure that includes various aspects of information
Ac
processing (such as perception, attention, working memory, etc.), which could be separated in further investigations by manipulating experimental paradigm. The cognitive workload could be considered akin to Lindsley’s local cortical activation in the EEG studies of mental activity [25,26]. During the performance of the ANT [22], the ratios between the late cue and the late target ERP components in the parietal and frontal areas were compared, in order to examine the regional differences in the processing of visuospatial information. We also estimated the
Page 6 of 21
5 relationship of these phenomena to spatial reasoning, by correlating the P3 with individuals’ scores on the Block Design subtest, which predominantly addresses spatial non-verbal abilities [27,28] together with the Vocabulary subtest (as a reference) from the Wechsler Intelligence Scale for Children (WISC-III) [29].
ip t
Material and Methods Subjects
cr
Twenty-three typically developing boys aged 10–13 years (11.3 ± 0.86) with no history of
us
neurological, psychiatric, or drug-related disorders and not being treated with psychotropics were submitted to a medical examination, to the K-SADS inventory [30], and to the DSM-
an
IVTR diagnosis for evaluation of ADHD. The Block Design and Vocabulary subtests of the WISC-III [29] were administered to estimate the intelligence quotient (I.Q.). The I.Q. scores
M
in typically developing children estimated by these two subtests showed a high correlation (r = 0.91) with the results from the full WISC scale [31]. Subjects with I.Q.< 80 were excluded.
ed
Each subject participated only once.
The study was conducted in accordance with the Code of Ethics of the World Medical
ce pt
Association and was approved by the Ethics Committee of our Institute. All the primary caregivers gave written informed consent. Attentional Network Test
Ac
The ANT is a forced two-choice test. Our ANT version was designed according to Kratz et al. [22]. The visual stimuli were displayed on the rectangular cyan background (25°×18°) on an LED monitor. The target signal was a yellow fish image (1.7°×1.1°), which equiprobably appeared 3.5° above or below the central fixation point (a black cross, 1.4°) between two identical fishes (distractors) on each side, 100 ms after their appearance. The target appeared for 350 ms. The horizontal orientation of the target to the right or to the left was equiprobable and the same (congruent) or opposite (incongruent) to the orientation of the
Page 7 of 21
6 distractors. The congruence was not taken into account. The random interval between the trials was 1 to 2 s. The cue signal was a red star (1.4°) that appeared for 150 ms, 1650 ms before the target appeared. There were three equiprobable cue conditions corresponding to this signal’s position or its non-appearance: 1) at the subsequent upper or lower position of the
ip t
target - Spatial cue condition; 2) at the central fixation point - Neutral cue condition; or 3) Nocue condition. All cues were valid. The target onset was not predicted in the No-cue condition
cr
only.
us
The subjects were instructed to visually fixate on the central cross and observe the horizontal orientation of a target stimulus. They were then told to immediately press with his
an
index or middle finger the left or right arrow key of the keyboard, according to the target horizontal orientation. They used the preferred hand in accordance with their handedness.
M
The test was organized in 9 blocks, each block consisting of 24 trials, i.e. 8 trials of each cue condition, presented in random order. The first block was used only for training. After
ed
each block, the system informed the subject about his performance, showing on the screen the mean reaction time (RT) and the accuracy (AC), i.e. the percentage of correct responses. Each
ce pt
block began when the subject pressed the space bar, indicating that he was ready. The test lasted about 20 minutes. For each cue condition, we estimated the average AC, RT and its standard deviation, termed the intraindividual variation (IV).
Ac
Participants who had AC < 70% were excluded, since their performances were considered to be random [22]. The speed-accuracy tradeoff was estimated as AC×RT. A value lower than the group mean minus two standard deviations was also considered a criterion for exclusion. EEG acquisition and ERP analysis During the ANT, the subject's EEG was recorded by a Nihon Kohden NK1200 System at 20 scalp points according to the International 10/20 system, with reference at the central leads (linked C3 and C4, the physical reference of the System) corresponded to the fronto-central
Page 8 of 21
7 [22] or central [18,32] references used in ANT studies. The impedance was below 10 kῼ. The sampling frequency was 1000 Hz. A trigger signal synchronized the target onsets with EEG. The EEG was filtered with a band pass of 0.5-150 Hz and suppression of 60 Hz noise. We manually eliminated artifacts. The EEG signals from the mid-frontal (Fz) and mid-parietal
ip t
(Pz) channels were averaged for the time window of 1650 ms before and 1000 ms after the trigger for all the cue conditions.
cr
We next selected the late components of ERPs from cues and targets within the time
us
window limited by the points where the average waves for each cue condition crossed the zero baseline (Fig. 1, A and B). For each subject, we estimated the absolute values of the P3 peak
an
and mean amplitudes (µV).
Each of these parameters in each derivation, as well as the AC, RT and IVRT were
M
statistically compared among all cue conditions. Results were considered statistically significant for p ≤ 0.05.
ed
Since most of the data did not show deviations in scedasticity according to Levene’s test (except for IVRT) and in normality according to the Shapiro-Wilk test (except for AC),
ce pt
parametric tests were applied because of their greater robustness [33]. These data also did not show sphericity deviations by Mauchly's test. This allowed us to apply the analysis of variance for repeated measures (r-ANOVA, with post-hoc pair-wise comparisons adjusted by the
Ac
Bonferroni method) to the data measured in the three cue conditions. For cue P3s, which were recorded only in the two cue conditions, the amplitude data were compared by the paired Student’s T-test, provided their intercorrelation by Pearson’s test was significant. The data for IVRT and AC were evaluated by the Friedman test followed by the Wilcoxon post-hoc pair test. Finally, we evaluated the correlations between the scores on the Block Design subtest and P3 characteristics by Pearson’s correlation coefficient, tracing linear regression models by the least-squares method, using the slope and intercept coefficients.
Page 9 of 21
8 Results Three subjects were excluded because of AC < 70%; therefore, the total number of subjects was 20 (5 left-handed). The behavioral data for each cue condition were similar to those of the previous
ip t
publication [22]. The RT for the No-, Neutral and Spatial cue conditions decreased respectively from 580 to 540 ms (p ≤ 0.019) (Table 1). The AC was about 90%, on average,
cr
with no difference among the cue conditions.
us
In Pz, the late positive components of the cue ERPs had two peaks (Fig. 1, upper). The first, larger peak had a latency of 300 ms, with amplitude 2.07 ± 2.73 µV for the Neutral and
an
4.16 ± 2.80 µV for the Spatial cue conditions. After the target onset, the second, larger peak appeared at 365 ms, with 6.82 ± 2.12, 7.78 ± 2.66 and 8.24 ± 2.57 µV for the No-, Neutral and
M
Spatial cue conditions, respectively). The latencies did not differ significantly between the conditions. The P3 magnitude for the Spatial cue was significantly higher than for the Neutral
ed
cue for both the mean amplitudes (p < 0.001) and the maximum peak amplitude (p = 0.032) (Table 2). The subsequent target P3 wave varied according to the cue condition, in the
ce pt
opposite way. The above, larger Spatial cue P3 (compared with the Neutral cue P3) was followed by the target P3 wave, which was smaller than those of the Neutral cue (p = 0.017 for mean amplitudes and p=0.011 for peak amplitude) and No-cue conditions (p =0.001 and p
Ac
= 0.005, respectively) (Table 2). Meanwhile, the target P3s for the Neutral cue and No-cue conditions were approximately equal. In Fz, the stimuli elicited similar negative late ERP complexes (negative because of the EEG reference to the central leads), with two peaks at 300 and 470 ms from the cue, and 370 and 490 ms from the target onsets (Fig. 1, lower). However, the magnitudes of these responses did not depend on the cue conditions. The sums of the cue and target mean or the maximum peak amplitudes of the late ERP
Page 10 of 21
9 components in Pz were almost equal for the Spatial and Neutral cue conditions (mean amplitudes: 3.07 ± 1.08 µV and 3.19 ± 1.36 µV, p=0.54; peak amplitudes: 15.01 ± 6.00 µV and 14.53 ± 6.15 µV, p = 0.86). These sums for the Spatial cueing showed a high correlation with the sums for the Neutral cueing (r = 0.83, p < 0.001, for mean amplitude; r = 0.74, p <
ip t
0.001 for peak amplitudes), demonstrating that this phenomenon was highly consistent for each individual subject.
cr
The scores on the Block Design subtest correlated with the parietal cue peak amplitudes in
us
the Neutral cue (r = 0.56, p = 0.008) and Spatial cue (r = 0.63, p = 0.002) conditions (Fig. 2); while the scores of the Vocabulary subtest as well as the target peak amplitudes showed no
with the scores of the WISC-III subtests.
M
Discussion
an
significant correlations with other characteristics. The frontal ERPs showed no correlation
Here, the parietal cue and target P3s proved to be sensitive to visuospatial processing,
ed
while at the frontal site, the analogue component showed no dependence on the character of the cueing. Spatial information in the cue signal not only augmented the amplitude of the cue
ce pt
P3 itself (in relation to the spatially neutral temporal cue), as observed in previous studies [14], but proportionally reduced the consequent target P3. The latter effect concords with one of the previous findings [16] and contrasts with the other [14]. These results suggest that this
Ac
component of the spatial cue ERP reflects visuospatial encoding, which remains in the shortterm memory and contributes to the subsequent "decision-making" process, consequently reducing the cognitive load on the subsequent target response. As a result, the sum of the mean or peak amplitudes of P3 in the pair “cue and target” for the Spatial cue condition was almost equal to that of the Neutral cue condition. This appears to show that the total load on the parietal networks during processing of a paired cue and target stimuli is similar in both cue conditions. The cognitive load refers to the total amount of mental effort being used in the
Page 11 of 21
10 short-term memory [23,24]. Polich considered the parietal P3 in response to relevant (target) stimuli in the OddBall paradigm to be related to short-term memory regarding the estimation of the upcoming stimuli [3,8]. The failure of the visually presented spatially neutral cues (containing only temporal information about the upcoming appearance of the target) to
ip t
influence the parietal target P3 points to a certain specificity of this short-term memory for spatial information processing. The similarity of the target P3 in the Neutral and No-cue
cr
conditions also suggests that the reduction in the magnitude of P3 observed in Spatial-cue
us
conditions was not related either to attentional processes or to those related to time encoding, but probably to the spatial information conveyed by the cue.
an
A particular relationship between the parietal P3 and visuospatial processing may also be evidenced by our results showing a significant positive correlation of the cue P3 with the
M
individual’s scores on the Block design subtest of WISC-III (Perceptual Reasoning Index scores) [29]), which addresses mainly non-verbal spatial skills and visuospatial abilities
ed
[27,28]. This finding indicates that this ERP component may, to some extent, reflect the individual’s visuospatial intelligence in general, and not only the current operational
ce pt
attentional processes related to decision-making. In this study, the parietal and frontal waveforms were similar to those described by Kratz et al. [22], who used almost the same experimental protocols and conditions. In contrast to the
Ac
traditional ANT protocol [18,21], these authors [22] set the target signal to appear 100 ms after the distractors, instead of appearing simultaneously. This approach may have caused some difference in the waveform from those observed by other investigators [18], because the target P3 may be superimposed on potentials related to distractors. Even so, the maximum peaks of the late ERP components observed obviously meet the criteria for the P3. The results obtained have revealed some additional functional aspects of the cognitive/attentional late ERP component P3, regarding visuospatial information processing in
Page 12 of 21
11 the parietal cortex. When the cue contains this information, the amplitudes of this component graphically demonstrate the allocation of the visuospatial processing load at the cue stage, with proportional relieving of the subsequent target stage. This effect was characteristic for spatial cueing and not for temporal cueing. This specificity was supported by the positive
ip t
correlation of parietal cue P3 with the individuals’ capacity for spatial reasoning. Although our results were obtained with preadolescents, we presume that they reflect
cr
universal phenomena since they accord with the literature data on the parietal cue and/or target
us
P3 amplitude in adults [16,18,19]. However, further investigation is needed to verify them in
an
adults.
Acknowledgements
M
The authors acknowledge the support of the Programa de Incentivo a Pesquisa (PIP) of
ed
the National Institute Fernandes Figueira (project IFF-008-Fio-13-3-2).
References
ce pt
[1] S. Sutton, M. Braren, J. Zubin, E.R. John, E.R., Evoked-Potential Correlates of Stimulus Uncertainty, Science, 150 (1965), 1187-1188. [2] T. Hruby, P. Marsalek, Event-related potentials - the P3 wave, Acta Neurobiol. Exp.
Ac
(Wars), 63 (2003) 55-63.
[3] J. Polich, Updating P300: an integrative theory of P3a and P3b, Clin. Neurophysiol. 118 (2007) 2128-2148. [4] T. W. Picton, The P300 wave of the human event-related potential, J. Clin. Neurophysiol. 4 (1992) 456-479. [5] M. Falkenstein, N.A. Koshlykova, V.N. Kiroj, J. Hoormann, J. Hohnsbein, Late ERP components in visual and auditory Go/Nogo tasks, Electroencephalogr. Clin. Neurophysiol.
Page 13 of 21
12 96 (1995) 36-43. [6] B. Kopp, S. Tabeling, C. Moschner, K. Wessel, Fractionating the neural mechanisms of cognitive control, J. Cogn. Neurosci. 18 (2006) 949-965. [7] R. Verleger, N. Baur, M. F. Metzner, K. Śmigasiewicz, The hard oddball: Effects of
ip t
difficult response selection on stimulus-related P3 and on response-related negative potentials, Psychophysiology 51(2014) 1089-1100.
cr
[8] J. Polich, J.R. Criado, Neuropsychology and neuropharmacology of P3a and P3b, Int.
us
J. Psychophysiol. 60 (2006) 172-185.
[9] D.E. Linden, The p300: where in the brain is it produced and what does it tell us?
an
Neuroscientist. 11 (2005) 563-576.
[10] L.G. Ungerleider, J.V. Haxby, 'What' and 'where' in the human brain, Curr. Opin.
M
Neurobiol. 4 (1994)157-165.
[11] D.J. Kravitz, K.S. Saleem, C.I. Baker, M. Mishkin, A new neural framework for
ed
visuospatial processing, Nat. Rev. Neurosci. 12(2011) 217-230. [12] A.R. Luria, The working brain: An introduction into neuropsychology, Basic Books,
ce pt
New York, 1973.
[13] J. Gottlieb, From thought to action: the parietal cortex as a bridge between perception, action, and cognition, Neuron 53 (2007) 9-16.
Ac
[14] M.J. Wright, G.M. Geffen, L.B. Geffen, Event related potentials during covert orientation of visual attention: effects of cue validity and directionality, Biol. Psychol. 41(1995) 183-202.
[15] M. Finke, C. Escera, F. Barceló, The effects of foreknowledge and task-set shifting as mirrored in cue- and target-locked event-related potentials, PLoS One 7(2012) e49486. [16] K. Hugdahl, H. Nordby, Electrophysiological correlates to cued attentional shifts in the visual and auditory modalities, Behav. Neural. Biol. 62(1994) 21-32.
Page 14 of 21
13 [17] E. Donchin, Presidential Address, 1980: Surprise!...Surprise?, Psychophysiology 18 (1981) 493-513. [18] A.H. Neuhaus, C. Urbanek, C. Opgen-Rhein, E. Hahn, T.M. Ta, S. Koehler, M. Gross, M. Dettling, Event-related potentials associated with Attention Network Test, Int. J.
ip t
Psychophysiol. 76 (2010) 72-79. [19] C. Perchet, L. García-Larrea, Visuospatial attention and motor reaction in children:
cr
an electrophysiological study of the "Posner" paradigm, Psychophysiology 37(2000) 231-241.
us
[20] J. Fan, M. Posner, Human attentional networks, Psychiatr Prax. 31(2004) Suppl. 2 S210-S214.
an
[21] J. Fan, B.D. McCandliss, T. Sommer, A. Raz, M.I. Posner, Testing the efficiency and independence of attentional networks, J. Cogn. Neurosci. 14(2002) 340-347
M
[22] O. Kratz, P. Studer, S. Malcherek, K. Erbe, G.H. Moll, H. Heinrich, Attentional processes in children with ADHD: An event-related potential study using the attention
ed
network test, Int. J. Psychophysiol. 81 (2011) 82–90. [23] J. Sweller, Cognitive Load During Problem Solving: Effects on Learning, Cogn. Sci.
ce pt
12 (1988) 257-285.
[24] J.B. Isreal, C.D. Wickens, G.L. Chesney, E. Donchin, The Event-Related Brain Potential as an Index of Display-Monitoring Workload, Hum. Factors 22 (1980) 211-224.
Ac
[25] D.B. Lindsley, Attention, consciousness, sleep and wakefulness, in: J. Field (Ed.), Handbook of Physiology, Neurophysiology, vol. 3, Thomas, Washington, 1960, pp. 1553– 1593.
[26] V.V. Lazarev, The relationship of theory and methodology in EEG studies of mental activity, Int. J. Psychophysiol. 62 (2006) 384–393. [27] A.S. Kaufman, Inteligence Testing with the WISC-III, Wiley, New York, 1994. [28] G. Groth-Marnat, Handbook of psychological assessment, 3rd ed., Wiley, New York,
Page 15 of 21
14 1999. [29] D. Wechsler, Wechsler Intelligence Scale for Children, 3rd Edition (WISC-III): Manual, San Antonio, The Psychological Corporation, 1991. [30] T. Matuschek, S. Jaeger, S. Stadelmann, et al., Implementing the K-SADS-PL as a
ip t
standard diagnostic tool: Effects on clinical diagnoses, Psychiatry Res. 28 (2016) 119-124. [31] C. B. Mello, N. Argollo, B. P. M. Shayer, N. Abreu, K. Godinho, P. Durán, F.
cr
Vargem, M. Muszkat, M. C. Miranda, O. F. A. Bueno, Abbreviated version of the WISC-III:
us
correlation between estimated IQ and global IQ of brazilian children, Psic.: Teor. e Pesq. 27 (2011) 149-155.
an
[32] D.A. Kaufman, C.N. Sozda, V.M. Dotson, W.M. Perlstein, An Event-Related Potential Investigation of the Effects of Age on Alerting, Orienting, and Executive Function,
M
Front. Aging Neurosci. 8 (2016) 99.
[33] P. Armitage, G. Beery, J.N.S. Matthews, Statistical methods in medical research. 4th
Ac
ce pt
ed
ed. Malden: Blackwell Science; 2002.
Page 16 of 21
15 Figure captions
Fig. 1. Grand event-related potentials (ERPs) averaged across subjects (n=20), evoked by cue and target stimuli in mid-parietal (Pz) and mid-frontal (Fz) leads. The cue (at left) and
ip t
target ERPs (at right) for the No-cue (light gray), Neutral cue (dotted gray), and Spatial cue (black) conditions. Dotted rectangles contain estimated components P3. Arrows mark the
us
cr
moments of the cue, flanks, and target presentations.
Fig. 2. Scatter plots showing the correlations between maximum peak amplitudes of cue
an
and target component P3 and scores on the Block Design subtest (WISC-III) in Neutral and
Ac
ce pt
ed
M
Spatial cue conditions. r – Pearson’s correlation coefficient.
Page 17 of 21
Ac ce p
te
d
M
an
us
cr
ip t
Figure Click here to download high resolution image
Page 18 of 21
Ac ce p
te
d
M
an
us
cr
ip t
Figure Click here to download high resolution image
Page 19 of 21
ip t
Table
Table 1
92,89 93,67 91,87 Mean (ms)
RT
NoNeutral Spatial
581,81 561,58 539,25 Mean (ms)
NoNeutral Spatial
162,72 134,01 163,49
ep te
IVRT
us
AC
NoNeutral Spatial
Standard Friedman Test(*) deviation (%) 9,47 Chi-Square = 0.53 8,52 p = 0.765 13,10 Standard r-ANOVA deviation (ms) 136,37 F-ratio = 13.25 125,54 p < 0.001 107,47 Standard Friedman deviation (ms) Test(**) 64,16 Chi-Square = 3.10 41,94 p = 0.212 99,84
an
Mean (%)
M
Cue condition
d
Variable
cr
Behavioral data
Difference, post-hoc test No difference Bonferroni method
No- & Neutral, p = 0.019 No- & Spatial, p < 0.001 Neutral & Spatial, p = 0.010
No difference
AC, accuracy, i.e. percentage of correct responses; RT, reaction time; IVRT, intraindividual variability of RT.
Ac c
(*) Distributions are not normal according to Shapiro-Wilk test: p < 0.001. (**) Variances are not equal according to Levene’s test: F-ratio = 5.89, p = 0.04.
Page 20 of 21
cr
ip t
Table
us
Table 2
Comparisons between cue and target P3 amplitudes for different cueing conditions, in Pz lead
Cue signal Target signal
Neutral cue No-cue No-cue Neutral cue
Mean amplitude (µV) 0.86 ± 0.95 Spatial cue 5.05 ± 1.52 Neutral cue 5.05 ± 1.52 Spatial cue 4.90 ± 1.88 Spatial cue Peak amplitude (µV) 4.07 ± 3.64 Spatial cue 10.87 ± 3.53 Neutral cue 10.87 ± 3.53 Spatial cue 10.45 ± 3.53 Spatial cue
Mean ± standard deviation
an
Condition
2.51 ± 1.52 4.90 ± 1.88 3.76 ± 1.74 3.76 ± 1.74
M
Target signal
Neutral cue No-cue No-cue Neutral cue
Mean ± standard deviation
d
Cue signal
Condition
ep te
Event
5.74 ± 3.99 10.45 ± 3.53 9.25 ± 2.51 9.25 ± 2.51
r-ANOVA r-ANOVA Significance (*) F-ratio p-value -
-
F = 173.85
p < 0.001
-
-
F = 217.2
p < 0.001
p < 0.001 p = 1.000 p = 0.001 p = 0.011 p = 0.032 p = 0.731 p = 0.005 p = 0.017
Ac c
(*) for cue signals, paired Student’s T-test; for target signals, post-hoc correction by Bonferroni method.
Page 21 of 21
S0304-3940(17)30251-3 http://dx.doi.org/doi:10.1016/j.neulet.2017.03.031 NSL 32719
To appear in:
Neuroscience Letters
Received date: Revised date: Accepted date:
9-9-2016 2-2-2017 17-3-2017
Please cite this article as: D.M. Abramov, M. Pontes, A.T. Pontes, C.A. MouraoJunior, J. Vieira, C.Q. Cunha, T. Tamborino, P.R. Galhanone, L.C. deAzevedo, V.V. Lazarev, Visuospatial information processing load and the ratio between parietal cue and target P3 amplitudes in the Attentional Network Test, Neuroscience Letters (2017), http://dx.doi.org/10.1016/j.neulet.2017.03.031 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Ac
ce
pt
ed
M
an
us
cr
i
Graphical Abstract
Page 1 of 21
*Highlights
Highlights - The cue containing spatial information about the target increases parietal cue P3. - This spatial cue reduces the subsequent parietal target P3. - The magnitudes of cue and target P3 total the same in spatial and neutral cueing.
Ac
ce pt
ed
M
an
us
cr
- The parietal P3 reflects visuospatial information processing.
ip t
- The parietal cue P3 positively correlates with the ability to perform spatial tasks.
Page 2 of 21
*Manuscript Click here to download Manuscript: Manuscript revised i2017.doc
Click here to view linked References
Visuospatial information processing load and the ratio between parietal cue and target P3 amplitudes in the Attentional Network Test Dimitri M. Abramova, Monique Pontesa, Adailton T. Pontesa, Carlos A. Mourao-Juniorb, Juliana Vieiraa, Carla Quero Cunhaa, Tiago Tamborinoa, Paulo R. Galhanonea, Leonardo C.
a
ip t
deAzevedoa, Vladimir V. Lazareva,*. Laboratory of Neurobiology and Clinical Neurophysiology, National Institute of Women,
cr
Children and Adolescents Health Fernandes Figueira, Oswaldo Cruz Foundation
(FIOCRUZ), Av. Ruy Barbosa, 716, Flamengo, Rio de Janeiro, RJ, 22250-020, Brazil. Department of Physiology, Federal University of Juiz de Fora, Campus UFJF - ICB, Juiz de
us
b
Corresponding author. E-mail address: [email protected] (V.V. Lazarev).
ed
M
*
an
Fora, MG, 36036-900, Brazil.
Abstract
ce pt
In ERP studies of cognitive processes during attentional tasks, the cue signals containing information about the target can increase the amplitude of the parietal cue P3 in relation to the ‘neutral’ temporal cue, and reduce the subsequent target P3 when this
Ac
information is valid, i.e. corresponds to the target’s attributes. The present study compared the cue-to-target P3 ratios in neutral and visuospatial cueing, in order to estimate the contribution of valid visuospatial information from the cue to target stages of the task performance, in terms of cognitive load. The P3 characteristics were also correlated with the results of individuals’ performance of the visuospatial tasks, in order to estimate the relationship of the observed ERP with spatial reasoning. In 20 typically developing boys, aged 10–13 years (11.3 ± 0.86), the intelligence quotient (I.Q.) was estimated by the Block Design and Vocabulary subtests from the WISC-
Page 3 of 21
2 III. The subjects performed the Attentional Network Test (ANT) accompanied by EEG recording. The cued two-choice task had three equiprobable cue conditions: No cue, with no information about the target; Neutral (temporal) cue, with an asterisk in the center of the visual field, predicting the target onset; and Spatial cues, with an asterisk in the upper or lower
ip t
hemifield, predicting the onset and corresponding location of the target. The ERPs were estimated for the mid-frontal (Fz) and mid-parietal (Pz) scalp derivations.
cr
In the Pz, the Neutral cue P3 had a lower amplitude than the Spatial cue P3; whereas
us
for the target ERPs, the P3 of the Neutral cue condition was larger than the Spatial cue P3. However, the sums of the magnitudes of the cue and target P3 were equal in the spatial and
an
neutral cueing, probably indicating that in both cases the equivalent information processing load is included in either the cue or the target reaction, respectively. Meantime, in the Fz, the
M
analogue ERP components for both the cue and target stimuli did not depend on the cue condition. The results show that, in the parietal site, the spatial cue P3 reflects the processing
ed
of visuospatial information regarding the target position. This contributes to the subsequent "decision-making", thus reducing the information processing load on the target response,
ce pt
which is probably reflected in the lower P3. This finding is consistent with the positive correlation of parietal cue P3 with the individual's ability to perform spatial tasks as scored by
Ac
the Block Design subtest.
Key words: visuospatial information processing; event related potentials; P3; cueing of target; parietal cortex; Attentional Network Test.
Page 4 of 21
3 Introduction The positive component of the event-related potential (ERP), which appears 300-500 ms after the event occurrence, is known as P3 or P300 [1-3]. This wave is related to a subjective estimation of the signal. It is usually elicited in goal-directed behavior during
ip t
attentional tasks [2,3], such as the OddBall [4], Go-NoGo [5], and other paradigms. It shows high sensitivity to the properties of a task, such as novelty, difficulty, and so on, and correlate
cr
with characteristics of attention and control [2,3,6,7].
us
The P3 is associated with the activity of temporo-parietal networks connected with the frontal cortex and dependent on the modality of a sensory stimulus [3,8,9]. The maximum
an
manifestation of this component is recorded in the parietal region [3] above the associative temporo-parieto-occipital cortex (TPO), which is responsible for the spatial processing of the
M
visual system [10,11], integration of information for action, and simultaneous spatial/quasispatial types of syntheses [11-13]. In general, these cortical areas relate to encoding, storage
ed
and retrieval of information [6,12].
ce pt
The cues of different modalities containing information about the target evoke higher amplitudes of the cue P3 [14,15] than the "neutral" temporal cue, which predicts the moment of the target's appearance. The valid cue signals corresponding to the target attributes can lead to reduction of the target P3 amplitude [16], which apparently reflects a higher probability of
Ac
the target appearing (according to [1,17]), and less subjective efforts in perceptual processing, attention orientation and control, probably determined by information that the cue conveyed. Importantly, this inverse relationship between the cue and target P3 is apparent when compared with the neutral temporal cueing. This follows from several studies, which, however, did not assess this effect directly, describing the cue or target P3 separately in the framework of other research objectives [18] or without neutral temporal cueing as a reference [16,19]. However, a temporally cued target P3 does not always show a difference from a
Page 5 of 21
4 validly cued target P3 [14]. This problem deserves special investigation, since the valid-cueing paradigm allows one to track the information-processing load from the cue to the target stage of the task performance. The present study evaluated the ratio between the cue and the target P3, based on the
ip t
Posner`s Attentional Network Test (ANT) [18,20-22]. The ANT operates with neutral temporal, valid spatial and no-cueing of the target onset and/or location in the visual field. We
cr
used this approach to detect the putative manifestation of the transfer of valid visuospatial
us
information from cue to target ERPs, and to estimate the contribution of this transfer to the final response in terms of cognitive load [17].
an
These terms refer to the total amount of mental efforts or neural resources being used during performance of the task [23]. The P3 is considered as a sensitive workload index
M
reflecting demand for information processing resources [17]. A likely reallocation of a portion of the workload from one task to a simultaneously executed another was shown by means of
ed
the P3 amplitude changes dependent on the first task’s difficulty and probably caused by a subject’s fixed capacity of neural resource available for the given type of mental operations
ce pt
[17,24]. We expected that the cue-to-target dynamics of the P3 amplitude in the ANT paradigm also could be adequately described by the workload. The latter might be a general and comprehensive psychophysiological measure that includes various aspects of information
Ac
processing (such as perception, attention, working memory, etc.), which could be separated in further investigations by manipulating experimental paradigm. The cognitive workload could be considered akin to Lindsley’s local cortical activation in the EEG studies of mental activity [25,26]. During the performance of the ANT [22], the ratios between the late cue and the late target ERP components in the parietal and frontal areas were compared, in order to examine the regional differences in the processing of visuospatial information. We also estimated the
Page 6 of 21
5 relationship of these phenomena to spatial reasoning, by correlating the P3 with individuals’ scores on the Block Design subtest, which predominantly addresses spatial non-verbal abilities [27,28] together with the Vocabulary subtest (as a reference) from the Wechsler Intelligence Scale for Children (WISC-III) [29].
ip t
Material and Methods Subjects
cr
Twenty-three typically developing boys aged 10–13 years (11.3 ± 0.86) with no history of
us
neurological, psychiatric, or drug-related disorders and not being treated with psychotropics were submitted to a medical examination, to the K-SADS inventory [30], and to the DSM-
an
IVTR diagnosis for evaluation of ADHD. The Block Design and Vocabulary subtests of the WISC-III [29] were administered to estimate the intelligence quotient (I.Q.). The I.Q. scores
M
in typically developing children estimated by these two subtests showed a high correlation (r = 0.91) with the results from the full WISC scale [31]. Subjects with I.Q.< 80 were excluded.
ed
Each subject participated only once.
The study was conducted in accordance with the Code of Ethics of the World Medical
ce pt
Association and was approved by the Ethics Committee of our Institute. All the primary caregivers gave written informed consent. Attentional Network Test
Ac
The ANT is a forced two-choice test. Our ANT version was designed according to Kratz et al. [22]. The visual stimuli were displayed on the rectangular cyan background (25°×18°) on an LED monitor. The target signal was a yellow fish image (1.7°×1.1°), which equiprobably appeared 3.5° above or below the central fixation point (a black cross, 1.4°) between two identical fishes (distractors) on each side, 100 ms after their appearance. The target appeared for 350 ms. The horizontal orientation of the target to the right or to the left was equiprobable and the same (congruent) or opposite (incongruent) to the orientation of the
Page 7 of 21
6 distractors. The congruence was not taken into account. The random interval between the trials was 1 to 2 s. The cue signal was a red star (1.4°) that appeared for 150 ms, 1650 ms before the target appeared. There were three equiprobable cue conditions corresponding to this signal’s position or its non-appearance: 1) at the subsequent upper or lower position of the
ip t
target - Spatial cue condition; 2) at the central fixation point - Neutral cue condition; or 3) Nocue condition. All cues were valid. The target onset was not predicted in the No-cue condition
cr
only.
us
The subjects were instructed to visually fixate on the central cross and observe the horizontal orientation of a target stimulus. They were then told to immediately press with his
an
index or middle finger the left or right arrow key of the keyboard, according to the target horizontal orientation. They used the preferred hand in accordance with their handedness.
M
The test was organized in 9 blocks, each block consisting of 24 trials, i.e. 8 trials of each cue condition, presented in random order. The first block was used only for training. After
ed
each block, the system informed the subject about his performance, showing on the screen the mean reaction time (RT) and the accuracy (AC), i.e. the percentage of correct responses. Each
ce pt
block began when the subject pressed the space bar, indicating that he was ready. The test lasted about 20 minutes. For each cue condition, we estimated the average AC, RT and its standard deviation, termed the intraindividual variation (IV).
Ac
Participants who had AC < 70% were excluded, since their performances were considered to be random [22]. The speed-accuracy tradeoff was estimated as AC×RT. A value lower than the group mean minus two standard deviations was also considered a criterion for exclusion. EEG acquisition and ERP analysis During the ANT, the subject's EEG was recorded by a Nihon Kohden NK1200 System at 20 scalp points according to the International 10/20 system, with reference at the central leads (linked C3 and C4, the physical reference of the System) corresponded to the fronto-central
Page 8 of 21
7 [22] or central [18,32] references used in ANT studies. The impedance was below 10 kῼ. The sampling frequency was 1000 Hz. A trigger signal synchronized the target onsets with EEG. The EEG was filtered with a band pass of 0.5-150 Hz and suppression of 60 Hz noise. We manually eliminated artifacts. The EEG signals from the mid-frontal (Fz) and mid-parietal
ip t
(Pz) channels were averaged for the time window of 1650 ms before and 1000 ms after the trigger for all the cue conditions.
cr
We next selected the late components of ERPs from cues and targets within the time
us
window limited by the points where the average waves for each cue condition crossed the zero baseline (Fig. 1, A and B). For each subject, we estimated the absolute values of the P3 peak
an
and mean amplitudes (µV).
Each of these parameters in each derivation, as well as the AC, RT and IVRT were
M
statistically compared among all cue conditions. Results were considered statistically significant for p ≤ 0.05.
ed
Since most of the data did not show deviations in scedasticity according to Levene’s test (except for IVRT) and in normality according to the Shapiro-Wilk test (except for AC),
ce pt
parametric tests were applied because of their greater robustness [33]. These data also did not show sphericity deviations by Mauchly's test. This allowed us to apply the analysis of variance for repeated measures (r-ANOVA, with post-hoc pair-wise comparisons adjusted by the
Ac
Bonferroni method) to the data measured in the three cue conditions. For cue P3s, which were recorded only in the two cue conditions, the amplitude data were compared by the paired Student’s T-test, provided their intercorrelation by Pearson’s test was significant. The data for IVRT and AC were evaluated by the Friedman test followed by the Wilcoxon post-hoc pair test. Finally, we evaluated the correlations between the scores on the Block Design subtest and P3 characteristics by Pearson’s correlation coefficient, tracing linear regression models by the least-squares method, using the slope and intercept coefficients.
Page 9 of 21
8 Results Three subjects were excluded because of AC < 70%; therefore, the total number of subjects was 20 (5 left-handed). The behavioral data for each cue condition were similar to those of the previous
ip t
publication [22]. The RT for the No-, Neutral and Spatial cue conditions decreased respectively from 580 to 540 ms (p ≤ 0.019) (Table 1). The AC was about 90%, on average,
cr
with no difference among the cue conditions.
us
In Pz, the late positive components of the cue ERPs had two peaks (Fig. 1, upper). The first, larger peak had a latency of 300 ms, with amplitude 2.07 ± 2.73 µV for the Neutral and
an
4.16 ± 2.80 µV for the Spatial cue conditions. After the target onset, the second, larger peak appeared at 365 ms, with 6.82 ± 2.12, 7.78 ± 2.66 and 8.24 ± 2.57 µV for the No-, Neutral and
M
Spatial cue conditions, respectively). The latencies did not differ significantly between the conditions. The P3 magnitude for the Spatial cue was significantly higher than for the Neutral
ed
cue for both the mean amplitudes (p < 0.001) and the maximum peak amplitude (p = 0.032) (Table 2). The subsequent target P3 wave varied according to the cue condition, in the
ce pt
opposite way. The above, larger Spatial cue P3 (compared with the Neutral cue P3) was followed by the target P3 wave, which was smaller than those of the Neutral cue (p = 0.017 for mean amplitudes and p=0.011 for peak amplitude) and No-cue conditions (p =0.001 and p
Ac
= 0.005, respectively) (Table 2). Meanwhile, the target P3s for the Neutral cue and No-cue conditions were approximately equal. In Fz, the stimuli elicited similar negative late ERP complexes (negative because of the EEG reference to the central leads), with two peaks at 300 and 470 ms from the cue, and 370 and 490 ms from the target onsets (Fig. 1, lower). However, the magnitudes of these responses did not depend on the cue conditions. The sums of the cue and target mean or the maximum peak amplitudes of the late ERP
Page 10 of 21
9 components in Pz were almost equal for the Spatial and Neutral cue conditions (mean amplitudes: 3.07 ± 1.08 µV and 3.19 ± 1.36 µV, p=0.54; peak amplitudes: 15.01 ± 6.00 µV and 14.53 ± 6.15 µV, p = 0.86). These sums for the Spatial cueing showed a high correlation with the sums for the Neutral cueing (r = 0.83, p < 0.001, for mean amplitude; r = 0.74, p <
ip t
0.001 for peak amplitudes), demonstrating that this phenomenon was highly consistent for each individual subject.
cr
The scores on the Block Design subtest correlated with the parietal cue peak amplitudes in
us
the Neutral cue (r = 0.56, p = 0.008) and Spatial cue (r = 0.63, p = 0.002) conditions (Fig. 2); while the scores of the Vocabulary subtest as well as the target peak amplitudes showed no
with the scores of the WISC-III subtests.
M
Discussion
an
significant correlations with other characteristics. The frontal ERPs showed no correlation
Here, the parietal cue and target P3s proved to be sensitive to visuospatial processing,
ed
while at the frontal site, the analogue component showed no dependence on the character of the cueing. Spatial information in the cue signal not only augmented the amplitude of the cue
ce pt
P3 itself (in relation to the spatially neutral temporal cue), as observed in previous studies [14], but proportionally reduced the consequent target P3. The latter effect concords with one of the previous findings [16] and contrasts with the other [14]. These results suggest that this
Ac
component of the spatial cue ERP reflects visuospatial encoding, which remains in the shortterm memory and contributes to the subsequent "decision-making" process, consequently reducing the cognitive load on the subsequent target response. As a result, the sum of the mean or peak amplitudes of P3 in the pair “cue and target” for the Spatial cue condition was almost equal to that of the Neutral cue condition. This appears to show that the total load on the parietal networks during processing of a paired cue and target stimuli is similar in both cue conditions. The cognitive load refers to the total amount of mental effort being used in the
Page 11 of 21
10 short-term memory [23,24]. Polich considered the parietal P3 in response to relevant (target) stimuli in the OddBall paradigm to be related to short-term memory regarding the estimation of the upcoming stimuli [3,8]. The failure of the visually presented spatially neutral cues (containing only temporal information about the upcoming appearance of the target) to
ip t
influence the parietal target P3 points to a certain specificity of this short-term memory for spatial information processing. The similarity of the target P3 in the Neutral and No-cue
cr
conditions also suggests that the reduction in the magnitude of P3 observed in Spatial-cue
us
conditions was not related either to attentional processes or to those related to time encoding, but probably to the spatial information conveyed by the cue.
an
A particular relationship between the parietal P3 and visuospatial processing may also be evidenced by our results showing a significant positive correlation of the cue P3 with the
M
individual’s scores on the Block design subtest of WISC-III (Perceptual Reasoning Index scores) [29]), which addresses mainly non-verbal spatial skills and visuospatial abilities
ed
[27,28]. This finding indicates that this ERP component may, to some extent, reflect the individual’s visuospatial intelligence in general, and not only the current operational
ce pt
attentional processes related to decision-making. In this study, the parietal and frontal waveforms were similar to those described by Kratz et al. [22], who used almost the same experimental protocols and conditions. In contrast to the
Ac
traditional ANT protocol [18,21], these authors [22] set the target signal to appear 100 ms after the distractors, instead of appearing simultaneously. This approach may have caused some difference in the waveform from those observed by other investigators [18], because the target P3 may be superimposed on potentials related to distractors. Even so, the maximum peaks of the late ERP components observed obviously meet the criteria for the P3. The results obtained have revealed some additional functional aspects of the cognitive/attentional late ERP component P3, regarding visuospatial information processing in
Page 12 of 21
11 the parietal cortex. When the cue contains this information, the amplitudes of this component graphically demonstrate the allocation of the visuospatial processing load at the cue stage, with proportional relieving of the subsequent target stage. This effect was characteristic for spatial cueing and not for temporal cueing. This specificity was supported by the positive
ip t
correlation of parietal cue P3 with the individuals’ capacity for spatial reasoning. Although our results were obtained with preadolescents, we presume that they reflect
cr
universal phenomena since they accord with the literature data on the parietal cue and/or target
us
P3 amplitude in adults [16,18,19]. However, further investigation is needed to verify them in
an
adults.
Acknowledgements
M
The authors acknowledge the support of the Programa de Incentivo a Pesquisa (PIP) of
ed
the National Institute Fernandes Figueira (project IFF-008-Fio-13-3-2).
References
ce pt
[1] S. Sutton, M. Braren, J. Zubin, E.R. John, E.R., Evoked-Potential Correlates of Stimulus Uncertainty, Science, 150 (1965), 1187-1188. [2] T. Hruby, P. Marsalek, Event-related potentials - the P3 wave, Acta Neurobiol. Exp.
Ac
(Wars), 63 (2003) 55-63.
[3] J. Polich, Updating P300: an integrative theory of P3a and P3b, Clin. Neurophysiol. 118 (2007) 2128-2148. [4] T. W. Picton, The P300 wave of the human event-related potential, J. Clin. Neurophysiol. 4 (1992) 456-479. [5] M. Falkenstein, N.A. Koshlykova, V.N. Kiroj, J. Hoormann, J. Hohnsbein, Late ERP components in visual and auditory Go/Nogo tasks, Electroencephalogr. Clin. Neurophysiol.
Page 13 of 21
12 96 (1995) 36-43. [6] B. Kopp, S. Tabeling, C. Moschner, K. Wessel, Fractionating the neural mechanisms of cognitive control, J. Cogn. Neurosci. 18 (2006) 949-965. [7] R. Verleger, N. Baur, M. F. Metzner, K. Śmigasiewicz, The hard oddball: Effects of
ip t
difficult response selection on stimulus-related P3 and on response-related negative potentials, Psychophysiology 51(2014) 1089-1100.
cr
[8] J. Polich, J.R. Criado, Neuropsychology and neuropharmacology of P3a and P3b, Int.
us
J. Psychophysiol. 60 (2006) 172-185.
[9] D.E. Linden, The p300: where in the brain is it produced and what does it tell us?
an
Neuroscientist. 11 (2005) 563-576.
[10] L.G. Ungerleider, J.V. Haxby, 'What' and 'where' in the human brain, Curr. Opin.
M
Neurobiol. 4 (1994)157-165.
[11] D.J. Kravitz, K.S. Saleem, C.I. Baker, M. Mishkin, A new neural framework for
ed
visuospatial processing, Nat. Rev. Neurosci. 12(2011) 217-230. [12] A.R. Luria, The working brain: An introduction into neuropsychology, Basic Books,
ce pt
New York, 1973.
[13] J. Gottlieb, From thought to action: the parietal cortex as a bridge between perception, action, and cognition, Neuron 53 (2007) 9-16.
Ac
[14] M.J. Wright, G.M. Geffen, L.B. Geffen, Event related potentials during covert orientation of visual attention: effects of cue validity and directionality, Biol. Psychol. 41(1995) 183-202.
[15] M. Finke, C. Escera, F. Barceló, The effects of foreknowledge and task-set shifting as mirrored in cue- and target-locked event-related potentials, PLoS One 7(2012) e49486. [16] K. Hugdahl, H. Nordby, Electrophysiological correlates to cued attentional shifts in the visual and auditory modalities, Behav. Neural. Biol. 62(1994) 21-32.
Page 14 of 21
13 [17] E. Donchin, Presidential Address, 1980: Surprise!...Surprise?, Psychophysiology 18 (1981) 493-513. [18] A.H. Neuhaus, C. Urbanek, C. Opgen-Rhein, E. Hahn, T.M. Ta, S. Koehler, M. Gross, M. Dettling, Event-related potentials associated with Attention Network Test, Int. J.
ip t
Psychophysiol. 76 (2010) 72-79. [19] C. Perchet, L. García-Larrea, Visuospatial attention and motor reaction in children:
cr
an electrophysiological study of the "Posner" paradigm, Psychophysiology 37(2000) 231-241.
us
[20] J. Fan, M. Posner, Human attentional networks, Psychiatr Prax. 31(2004) Suppl. 2 S210-S214.
an
[21] J. Fan, B.D. McCandliss, T. Sommer, A. Raz, M.I. Posner, Testing the efficiency and independence of attentional networks, J. Cogn. Neurosci. 14(2002) 340-347
M
[22] O. Kratz, P. Studer, S. Malcherek, K. Erbe, G.H. Moll, H. Heinrich, Attentional processes in children with ADHD: An event-related potential study using the attention
ed
network test, Int. J. Psychophysiol. 81 (2011) 82–90. [23] J. Sweller, Cognitive Load During Problem Solving: Effects on Learning, Cogn. Sci.
ce pt
12 (1988) 257-285.
[24] J.B. Isreal, C.D. Wickens, G.L. Chesney, E. Donchin, The Event-Related Brain Potential as an Index of Display-Monitoring Workload, Hum. Factors 22 (1980) 211-224.
Ac
[25] D.B. Lindsley, Attention, consciousness, sleep and wakefulness, in: J. Field (Ed.), Handbook of Physiology, Neurophysiology, vol. 3, Thomas, Washington, 1960, pp. 1553– 1593.
[26] V.V. Lazarev, The relationship of theory and methodology in EEG studies of mental activity, Int. J. Psychophysiol. 62 (2006) 384–393. [27] A.S. Kaufman, Inteligence Testing with the WISC-III, Wiley, New York, 1994. [28] G. Groth-Marnat, Handbook of psychological assessment, 3rd ed., Wiley, New York,
Page 15 of 21
14 1999. [29] D. Wechsler, Wechsler Intelligence Scale for Children, 3rd Edition (WISC-III): Manual, San Antonio, The Psychological Corporation, 1991. [30] T. Matuschek, S. Jaeger, S. Stadelmann, et al., Implementing the K-SADS-PL as a
ip t
standard diagnostic tool: Effects on clinical diagnoses, Psychiatry Res. 28 (2016) 119-124. [31] C. B. Mello, N. Argollo, B. P. M. Shayer, N. Abreu, K. Godinho, P. Durán, F.
cr
Vargem, M. Muszkat, M. C. Miranda, O. F. A. Bueno, Abbreviated version of the WISC-III:
us
correlation between estimated IQ and global IQ of brazilian children, Psic.: Teor. e Pesq. 27 (2011) 149-155.
an
[32] D.A. Kaufman, C.N. Sozda, V.M. Dotson, W.M. Perlstein, An Event-Related Potential Investigation of the Effects of Age on Alerting, Orienting, and Executive Function,
M
Front. Aging Neurosci. 8 (2016) 99.
[33] P. Armitage, G. Beery, J.N.S. Matthews, Statistical methods in medical research. 4th
Ac
ce pt
ed
ed. Malden: Blackwell Science; 2002.
Page 16 of 21
15 Figure captions
Fig. 1. Grand event-related potentials (ERPs) averaged across subjects (n=20), evoked by cue and target stimuli in mid-parietal (Pz) and mid-frontal (Fz) leads. The cue (at left) and
ip t
target ERPs (at right) for the No-cue (light gray), Neutral cue (dotted gray), and Spatial cue (black) conditions. Dotted rectangles contain estimated components P3. Arrows mark the
us
cr
moments of the cue, flanks, and target presentations.
Fig. 2. Scatter plots showing the correlations between maximum peak amplitudes of cue
an
and target component P3 and scores on the Block Design subtest (WISC-III) in Neutral and
Ac
ce pt
ed
M
Spatial cue conditions. r – Pearson’s correlation coefficient.
Page 17 of 21
Ac ce p
te
d
M
an
us
cr
ip t
Figure Click here to download high resolution image
Page 18 of 21
Ac ce p
te
d
M
an
us
cr
ip t
Figure Click here to download high resolution image
Page 19 of 21
ip t
Table
Table 1
92,89 93,67 91,87 Mean (ms)
RT
NoNeutral Spatial
581,81 561,58 539,25 Mean (ms)
NoNeutral Spatial
162,72 134,01 163,49
ep te
IVRT
us
AC
NoNeutral Spatial
Standard Friedman Test(*) deviation (%) 9,47 Chi-Square = 0.53 8,52 p = 0.765 13,10 Standard r-ANOVA deviation (ms) 136,37 F-ratio = 13.25 125,54 p < 0.001 107,47 Standard Friedman deviation (ms) Test(**) 64,16 Chi-Square = 3.10 41,94 p = 0.212 99,84
an
Mean (%)
M
Cue condition
d
Variable
cr
Behavioral data
Difference, post-hoc test No difference Bonferroni method
No- & Neutral, p = 0.019 No- & Spatial, p < 0.001 Neutral & Spatial, p = 0.010
No difference
AC, accuracy, i.e. percentage of correct responses; RT, reaction time; IVRT, intraindividual variability of RT.
Ac c
(*) Distributions are not normal according to Shapiro-Wilk test: p < 0.001. (**) Variances are not equal according to Levene’s test: F-ratio = 5.89, p = 0.04.
Page 20 of 21
cr
ip t
Table
us
Table 2
Comparisons between cue and target P3 amplitudes for different cueing conditions, in Pz lead
Cue signal Target signal
Neutral cue No-cue No-cue Neutral cue
Mean amplitude (µV) 0.86 ± 0.95 Spatial cue 5.05 ± 1.52 Neutral cue 5.05 ± 1.52 Spatial cue 4.90 ± 1.88 Spatial cue Peak amplitude (µV) 4.07 ± 3.64 Spatial cue 10.87 ± 3.53 Neutral cue 10.87 ± 3.53 Spatial cue 10.45 ± 3.53 Spatial cue
Mean ± standard deviation
an
Condition
2.51 ± 1.52 4.90 ± 1.88 3.76 ± 1.74 3.76 ± 1.74
M
Target signal
Neutral cue No-cue No-cue Neutral cue
Mean ± standard deviation
d
Cue signal
Condition
ep te
Event
5.74 ± 3.99 10.45 ± 3.53 9.25 ± 2.51 9.25 ± 2.51
r-ANOVA r-ANOVA Significance (*) F-ratio p-value -
-
F = 173.85
p < 0.001
-
-
F = 217.2
p < 0.001
p < 0.001 p = 1.000 p = 0.001 p = 0.011 p = 0.032 p = 0.731 p = 0.005 p = 0.017
Ac c
(*) for cue signals, paired Student’s T-test; for target signals, post-hoc correction by Bonferroni method.
Page 21 of 21