Brain Topogr DOI 10.1007/s10548-013-0340-8
BRIEF COMMUNICATION
Magnetic Source Localization of Early Visual Mismatch Response Ana Susac • Dirk J. Heslenfeld • Ralph Huonker Selma Supek
•
Received: 23 May 2013 / Accepted: 29 November 2013 Ó Springer Science+Business Media New York 2013
Abstract Previous studies have reported a visual analogue of the auditory mismatch negativity (MMN) response that is based on sensory memory. The neural generators and attention dependence of the visual MMN (vMMN) still remain unclear. We used magnetoencephalography (MEG) and spatio-temporal source localization to determine the generators of the sensory-memory-based vMMN response to non-attended deviants. Ten participants were asked to discriminate between odd and even digits presented at the center of the visual field while grating patterns with different spatial frequencies were presented outside the focus of attention. vMMN was calculated as the difference between MEG responses to infrequent gratings in oddball blocks and the same gratings in equiprobable blocks. The
peak latency of the vMMN response was between 100 and 160 ms. The neuromagnetic sources of the vMMN localized in the occipital cortex differed from the sources evoked by the equiprobable gratings and were stimulusdependent. Our results suggest the existence of separate neural systems for pre-attentive memory-based detection of visual change and provide new evidence that the vMMN is feature-specific. Keywords Visual mismatch negativity (vMMN) Magnetoencephalography (MEG) Source localization Deviance detection Occipital cortex
Introduction Electronic supplementary material The online version of this article (doi:10.1007/s10548-013-0340-8) contains supplementary material, which is available to authorized users. A. Susac (&) S. Supek Department of Physics, Faculty of Science, University of Zagreb, Bijenicka 32, 10 000 Zagreb, Croatia e-mail: [email protected] S. Supek e-mail: [email protected] A. Susac Oxford Centre for Human Brain Activity, University of Oxford, Oxford, UK D. J. Heslenfeld Department of Psychology, VU University, Amsterdam, The Netherlands e-mail: [email protected] R. Huonker Department of Neurology, Biomagnetic Center, Friedrich Schiller University of Jena, Jena, Germany e-mail: [email protected]
Detection of changes in the environment is crucial for adaptive behavior. In the auditory modality, the mismatch negativity (MMN) response in electroencephalographic (EEG) and magnetoencephalographic (MEG) recordings reflects pre-attentive detection of unexpected (deviant) stimuli in a sequence of frequent (standard) stimuli. The underlying mechanism of the MMN is thought to be a comparison of the deviant stimulus with a sensory-memory trace of the standard stimulus (Na¨a¨ta¨nen 1992). Auditory MMN has been studied for more than three decades (Na¨a¨ta¨nen et al. 1978). More recently, a visual analogue of the MMN has been reported (Heslenfeld 2003; Kimura et al. 2009; for recent reviews see Kimura 2012; Winkler and Czigler 2012). According to recent models, it seems that both MMN and vMMN reflect a prediction error which occurs when the actual stimulus is incongruent with the stimulus predicted on the basis of extracted regularities (Garrido et al. 2009; Kimura 2012; Winkler and Czigler 2012).
123
Brain Topogr
To consider an EEG/MEG response to deviant visual stimuli as a visual MMN (vMMN), several criteria should be met. First, physical differences between standard and deviant stimuli should be controlled. This can be done by comparing EEG/MEG responses to the same physical stimulus in both standard and deviant roles. Second, deviants should not be attended. This can be obtained by using a difficult visual task, unrelated to the deviants (Heslenfeld 2003). Third, to test the memory-based origin of the change detection, an equiprobable sequence should also be used (Schro¨ger and Wolff 1996). The ‘‘genuine’’ vMMN is then calculated as the difference between responses elicited by a deviant in the oddball sequence and the same stimulus in the equiprobable sequence (Czigler and Balazs 2002). Previous studies found a memory-based vMMN using an equiprobable sequence in addition to a standard oddball sequence (Astikainen et al. 2008; Czigler and Balazs 2002; Kimura et al. 2009). However, the neural generators of this ‘‘genuine’’ vMMN were explored only in one EEG study, in which attentional effects were not excluded (Kimura et al. 2010). We used MEG and spatio-temporal source localization to determine the generators of the genuine vMMN to unattended visual deviants.
(a)
Fig. 1 a MEG responses of four subjects to the same low SF grating presented as deviant (in the oddball sequence; thin blue line) and equiprobable stimulus (thin dashed red line). vMMN is calculated as the difference between these two responses (thick green line). MEG channels with the largest vMMN response are shown. Current dipoles
123
Methods Detailed methods are provided as electronic supplementary material. Evoked magnetic fields were measured in a magnetically shielded room at the Biomagnetic Center, Friedrich Schiller University of Jena, Germany using a 306-channel MEG system (Elekta Neuromag Oy, Helsinki, Finland). 24 letters (p = 0.8), 4 even digits (p = 0.1) and 4 odd digits (p = 0.1) were presented at a high rate (2 Hz) at the center of the screen for a duration of 100 ms. Participants were instructed to press a button for odd digits (Heslenfeld and Kurpershoek 2006). Concurrently with letters and digits, task-irrelevant gratings were presented outside the focus of attention. The gratings had an annular shape and consisted of a vertical black and white square wave pattern of either a low (1.5 cpd) or high (6 cpd) spatial frequency (SF). In the first oddball sequence, the high SF grating was used as deviant stimulus (p = 0.2), and the low SF as standard (p = 0.8). In the second oddball sequence, the standard and the deviant were reversed. In the equiprobable sequence, two gratings from the oddball sequences and three gratings with intermediate SFs (2, 3, 4.5 cpd) were presented with the equal
(b)
identified for the vMMN response in the 30-ms time window around participant’s vMMN peak latency are marked by arrows. Magnetic field maps are shown for the latency with the largest amplitude of the identified ECD. b Same as a but for high SF gratings (Color figure online)
Brain Topogr
probability (p = 0.2). We evaluated the ‘‘genuine’’ vMMN (Czigler and Balazs 2002) as the difference between MEG responses to the deviant in the oddball blocks and the same stimulus in the equiprobable blocks.
Results Differences in the evoked responses to the same stimulus presented as a deviant or as equiprobable stimulus were found in all subjects (Fig. 1). There was an initial deflection at about 80 ms, followed by the vMMN peaking between 100 and 160 ms. In addition, the vMMN response had opposite polarities for the low and high SF gratings (Fig. 1b). MEG responses to the low and high SF gratings presented in the equiprobable sequence were rather similar. When the low and high SF gratings were presented as deviants in the oddball sequence, the MEG responses diverged in the opposite direction from the corresponding equiprobable responses. The vMMN responses were localized in the occipital cortex for all participants (Fig. 2). Note that vMMN sources had the opposite orientation for low and high SF gratings (Figs. 1, 2). To compare the sources identified for the vMMN and the equiprobable stimulus for both gratings,
we submitted the source peak latencies and amplitudes to ANOVA. For peak source latencies, only a main effect of type of response was found (vMMN: 127 ± 9 ms, equiprobable: 119 ± 12 ms, F(1,9) = 5.81, p \ 0.05). Also for source amplitudes only a main effect of the type of response was found (vMMN: 19 ± 3 nAm, equiprobable: 41 ± 11 nAm, F(1,9) = 7.87, p \ 0.05). Finally, the mean distance and standard error between vMMN and equiprobable sources across subjects was 24 ± 4 mm for the low SF grating, and 23 ± 5 mm for the high SF grating, and it was larger than the mean distance between equiprobable sources for the low and high SF gratings (8 ± 3 mm, F(2,9) = 9.51, p \ 0.002).
Discussion To our knowledge, this is the first MEG study on the vMMN which used an equiprobable sequence in addition to the standard oddball sequence. We found different neuromagnetic responses to the same visual stimulus when presented in oddball and equiprobable blocks, in agreement with previous EEG studies (Astikainen et al. 2008; Czigler and Balazs 2002; Kimura et al. 2009). The first deflection at 80 ms in the difference waveform probably reflects a
Fig. 2 Locations of ECD for the vMMN (left) and equiprobable stimulus (right) identified for participant P1 in the time window 100–130 ms for the low SF grating (upper row) and the high SF grating (lower row)
123
Brain Topogr
sensory, refractoriness-based deviance detection system (Heslenfeld 2003), possibly a modulated N1m. The second, much larger deflection that appeared within 160 ms after stimulus onset, the vMMN, reflects a cognitive, predictionerror/memory-based deviance detection system (Kimura 2012). The source of the vMMN was located in the occipital lobe for all participants. This indicates that pre-attentive change detection is a relatively low-level, modality-specific process in the visual cortex. In the present study, we used a single dipole model to estimate vMMN sources for the bipolar field patterns around vMMN peak latency. Size of the stimuli, visual field location, and the use of differential topographies had probably caused cancellation of large parts of the magnetic field on the head surface (Josef Golubic et al. 2011). The equivalent single dipole accounted well for the data in both a visual and statistical sense (e.g., for most subjects g value was greater than 85 %) and most likely reflected contributions of the dominant, closely spaced occipital and occipito-parietal sources. On average, the location of the vMMN sources was 2 cm away from the sources for equiprobable stimulus localized in the same time window. Auditory MEG studies also showed separate generators for the N1 and MMN (Leva¨nen et al. 1996). Our spatio-temporal source localization study supports this finding in the visual modality. Our results are also consistent with the previous EEG and MEG studies that reported deviance-related activity in the middle occipital gyrus and cuneus (Kimura et al. 2010; Urakawa et al. 2010). However, neither of the two previous studies on the localization of the vMMN generators controlled for attention, as the tasks were the detection of the deviant stimuli (Kimura et al. 2010) or watching a silent video (Urakawa et al. 2010). In the present study, we used a task which was designed to control better for attentional effects (Heslenfeld and Kurpershoek 2006). Finally, vMMN responses for gratings with different SF’s depended on the physical features of the stimulus, probably reflecting differential processing of high SF information by parvocellular and low SF information by magnocellular visual channels (Tapia and Breitmeyer 2011). Auditory MMN was also reported to be stimulus dependent (Na¨a¨ta¨nen 1992), and Sulykos and Czigler (2011) reported a feature-related vMMN. Our data provide new evidence that stimulus-dependent mechanisms are underlying the vMMN as well, indicating that the vMMN is not caused by a generic mechanism sensitive to regularity violation as such.
123
Acknowledgments This study was supported by the European Union (COST B27 Action) and Croatian Ministry of Science, Education, and Sport (Grant 119-1081870-1252).
References Astikainen P, Lillstrang E, Ruusuvirta T (2008) Visual mismatch negativity for changes in orientation—sensory memory-dependent response. Eur J Neurosci 28:2319–2324 Czigler I, Balazs Winkler I (2002) Memory-based detection of taskirrelevant visual changes. Psychophysiology 39:869–873 Garrido MI, Kilner J, Stephan KE, Friston KJ (2009) The mismatch negativity: a review of underlying mechanisms. Clin Neurophysiol 120:453–463 Heslenfeld DJ (2003) Visual mismatch negativity. In: Polich J (ed) Detection of change: event-related potential and fMRI findings. Kluwer Academic Publishers, Boston, pp 41–59 Heslenfeld DJ, Kurpershoek CE (2006) Pre-attentive change detection in vision. 12th Annual Meeting of the Organization for Human Brain Mapping, Neuroimage Special Issue S30 Josef Golubic S, Susac A, Grilj V, Ranken D, Huonker R, Haueisen J, Supek S (2011) Size matters: MEG empirical and simulation study on source localization of the earliest visual activity in the occipital cortex. Med Biol Eng Comput 49:545–554 Kimura M (2012) Visual mismatch negativity and unintentional temporal-context-based prediction in vision. Int J Psychophysiol 83:144–155 Kimura M, Katayama J, Ohira H, Schro¨ger E (2009) Visual mismatch negativity: new evidence from the equiprobable paradigm. Psychophysiology 46:402–409 Kimura M, Ohira H, Schro¨ger E et al (2010) Localizing sensory and cognitive systems for pre-attentive visual deviance detection: an sLORETA analysis of the data of Kimura et al. (2009). Neurosci Lett 485:198–203 Leva¨nen S, Ahonen A, Hari R, McEvoy L, Sams M (1996) Deviant auditory stimuli activate human left and right auditory cortex differently. Cereb Cortex 6:288–296 Na¨a¨ta¨nen R (1992) Attention and brain function. Lawrence Erlbaum Associates, Hillsdale Na¨a¨ta¨nen R, Gaillard AWK, Ma¨ntysalo S (1978) Early selectiveattention effect on evoked potential reinterpreted. Acta Psychol 42:313–329 Schro¨ger E, Wolff C (1996) Mismatch response of the human brain to changes in sound location. Neuroreport 7:3005–3008 Sulykos I, Czigler I (2011) One plus one is less than two: visual features elicit non-additive mismatch-related brain activity. Brain Res 1398:64–71 Tapia E, Breitmeyer BG (2011) Visual consciousness revisited: magnocellular and parvocellular contributions to conscious and nonconscious vision. Psychol Sci 22:934–942 Urakawa T, Inui K, Yamashiro K, Tanaka E, Kakigi R (2010) Cortical dynamics of visual change detection based on sensory memory. Neuroimage 52:302–308 Winkler I, Czigler I (2012) Evidence from auditory and visual eventrelated potential (ERP) studies of deviance detection (MMN and vMMN) linking predictive coding theories and perceptual object representations. Int J Psychophysiol 83:132–143
BRIEF COMMUNICATION
Magnetic Source Localization of Early Visual Mismatch Response Ana Susac • Dirk J. Heslenfeld • Ralph Huonker Selma Supek
•
Received: 23 May 2013 / Accepted: 29 November 2013 Ó Springer Science+Business Media New York 2013
Abstract Previous studies have reported a visual analogue of the auditory mismatch negativity (MMN) response that is based on sensory memory. The neural generators and attention dependence of the visual MMN (vMMN) still remain unclear. We used magnetoencephalography (MEG) and spatio-temporal source localization to determine the generators of the sensory-memory-based vMMN response to non-attended deviants. Ten participants were asked to discriminate between odd and even digits presented at the center of the visual field while grating patterns with different spatial frequencies were presented outside the focus of attention. vMMN was calculated as the difference between MEG responses to infrequent gratings in oddball blocks and the same gratings in equiprobable blocks. The
peak latency of the vMMN response was between 100 and 160 ms. The neuromagnetic sources of the vMMN localized in the occipital cortex differed from the sources evoked by the equiprobable gratings and were stimulusdependent. Our results suggest the existence of separate neural systems for pre-attentive memory-based detection of visual change and provide new evidence that the vMMN is feature-specific. Keywords Visual mismatch negativity (vMMN) Magnetoencephalography (MEG) Source localization Deviance detection Occipital cortex
Introduction Electronic supplementary material The online version of this article (doi:10.1007/s10548-013-0340-8) contains supplementary material, which is available to authorized users. A. Susac (&) S. Supek Department of Physics, Faculty of Science, University of Zagreb, Bijenicka 32, 10 000 Zagreb, Croatia e-mail: [email protected] S. Supek e-mail: [email protected] A. Susac Oxford Centre for Human Brain Activity, University of Oxford, Oxford, UK D. J. Heslenfeld Department of Psychology, VU University, Amsterdam, The Netherlands e-mail: [email protected] R. Huonker Department of Neurology, Biomagnetic Center, Friedrich Schiller University of Jena, Jena, Germany e-mail: [email protected]
Detection of changes in the environment is crucial for adaptive behavior. In the auditory modality, the mismatch negativity (MMN) response in electroencephalographic (EEG) and magnetoencephalographic (MEG) recordings reflects pre-attentive detection of unexpected (deviant) stimuli in a sequence of frequent (standard) stimuli. The underlying mechanism of the MMN is thought to be a comparison of the deviant stimulus with a sensory-memory trace of the standard stimulus (Na¨a¨ta¨nen 1992). Auditory MMN has been studied for more than three decades (Na¨a¨ta¨nen et al. 1978). More recently, a visual analogue of the MMN has been reported (Heslenfeld 2003; Kimura et al. 2009; for recent reviews see Kimura 2012; Winkler and Czigler 2012). According to recent models, it seems that both MMN and vMMN reflect a prediction error which occurs when the actual stimulus is incongruent with the stimulus predicted on the basis of extracted regularities (Garrido et al. 2009; Kimura 2012; Winkler and Czigler 2012).
123
Brain Topogr
To consider an EEG/MEG response to deviant visual stimuli as a visual MMN (vMMN), several criteria should be met. First, physical differences between standard and deviant stimuli should be controlled. This can be done by comparing EEG/MEG responses to the same physical stimulus in both standard and deviant roles. Second, deviants should not be attended. This can be obtained by using a difficult visual task, unrelated to the deviants (Heslenfeld 2003). Third, to test the memory-based origin of the change detection, an equiprobable sequence should also be used (Schro¨ger and Wolff 1996). The ‘‘genuine’’ vMMN is then calculated as the difference between responses elicited by a deviant in the oddball sequence and the same stimulus in the equiprobable sequence (Czigler and Balazs 2002). Previous studies found a memory-based vMMN using an equiprobable sequence in addition to a standard oddball sequence (Astikainen et al. 2008; Czigler and Balazs 2002; Kimura et al. 2009). However, the neural generators of this ‘‘genuine’’ vMMN were explored only in one EEG study, in which attentional effects were not excluded (Kimura et al. 2010). We used MEG and spatio-temporal source localization to determine the generators of the genuine vMMN to unattended visual deviants.
(a)
Fig. 1 a MEG responses of four subjects to the same low SF grating presented as deviant (in the oddball sequence; thin blue line) and equiprobable stimulus (thin dashed red line). vMMN is calculated as the difference between these two responses (thick green line). MEG channels with the largest vMMN response are shown. Current dipoles
123
Methods Detailed methods are provided as electronic supplementary material. Evoked magnetic fields were measured in a magnetically shielded room at the Biomagnetic Center, Friedrich Schiller University of Jena, Germany using a 306-channel MEG system (Elekta Neuromag Oy, Helsinki, Finland). 24 letters (p = 0.8), 4 even digits (p = 0.1) and 4 odd digits (p = 0.1) were presented at a high rate (2 Hz) at the center of the screen for a duration of 100 ms. Participants were instructed to press a button for odd digits (Heslenfeld and Kurpershoek 2006). Concurrently with letters and digits, task-irrelevant gratings were presented outside the focus of attention. The gratings had an annular shape and consisted of a vertical black and white square wave pattern of either a low (1.5 cpd) or high (6 cpd) spatial frequency (SF). In the first oddball sequence, the high SF grating was used as deviant stimulus (p = 0.2), and the low SF as standard (p = 0.8). In the second oddball sequence, the standard and the deviant were reversed. In the equiprobable sequence, two gratings from the oddball sequences and three gratings with intermediate SFs (2, 3, 4.5 cpd) were presented with the equal
(b)
identified for the vMMN response in the 30-ms time window around participant’s vMMN peak latency are marked by arrows. Magnetic field maps are shown for the latency with the largest amplitude of the identified ECD. b Same as a but for high SF gratings (Color figure online)
Brain Topogr
probability (p = 0.2). We evaluated the ‘‘genuine’’ vMMN (Czigler and Balazs 2002) as the difference between MEG responses to the deviant in the oddball blocks and the same stimulus in the equiprobable blocks.
Results Differences in the evoked responses to the same stimulus presented as a deviant or as equiprobable stimulus were found in all subjects (Fig. 1). There was an initial deflection at about 80 ms, followed by the vMMN peaking between 100 and 160 ms. In addition, the vMMN response had opposite polarities for the low and high SF gratings (Fig. 1b). MEG responses to the low and high SF gratings presented in the equiprobable sequence were rather similar. When the low and high SF gratings were presented as deviants in the oddball sequence, the MEG responses diverged in the opposite direction from the corresponding equiprobable responses. The vMMN responses were localized in the occipital cortex for all participants (Fig. 2). Note that vMMN sources had the opposite orientation for low and high SF gratings (Figs. 1, 2). To compare the sources identified for the vMMN and the equiprobable stimulus for both gratings,
we submitted the source peak latencies and amplitudes to ANOVA. For peak source latencies, only a main effect of type of response was found (vMMN: 127 ± 9 ms, equiprobable: 119 ± 12 ms, F(1,9) = 5.81, p \ 0.05). Also for source amplitudes only a main effect of the type of response was found (vMMN: 19 ± 3 nAm, equiprobable: 41 ± 11 nAm, F(1,9) = 7.87, p \ 0.05). Finally, the mean distance and standard error between vMMN and equiprobable sources across subjects was 24 ± 4 mm for the low SF grating, and 23 ± 5 mm for the high SF grating, and it was larger than the mean distance between equiprobable sources for the low and high SF gratings (8 ± 3 mm, F(2,9) = 9.51, p \ 0.002).
Discussion To our knowledge, this is the first MEG study on the vMMN which used an equiprobable sequence in addition to the standard oddball sequence. We found different neuromagnetic responses to the same visual stimulus when presented in oddball and equiprobable blocks, in agreement with previous EEG studies (Astikainen et al. 2008; Czigler and Balazs 2002; Kimura et al. 2009). The first deflection at 80 ms in the difference waveform probably reflects a
Fig. 2 Locations of ECD for the vMMN (left) and equiprobable stimulus (right) identified for participant P1 in the time window 100–130 ms for the low SF grating (upper row) and the high SF grating (lower row)
123
Brain Topogr
sensory, refractoriness-based deviance detection system (Heslenfeld 2003), possibly a modulated N1m. The second, much larger deflection that appeared within 160 ms after stimulus onset, the vMMN, reflects a cognitive, predictionerror/memory-based deviance detection system (Kimura 2012). The source of the vMMN was located in the occipital lobe for all participants. This indicates that pre-attentive change detection is a relatively low-level, modality-specific process in the visual cortex. In the present study, we used a single dipole model to estimate vMMN sources for the bipolar field patterns around vMMN peak latency. Size of the stimuli, visual field location, and the use of differential topographies had probably caused cancellation of large parts of the magnetic field on the head surface (Josef Golubic et al. 2011). The equivalent single dipole accounted well for the data in both a visual and statistical sense (e.g., for most subjects g value was greater than 85 %) and most likely reflected contributions of the dominant, closely spaced occipital and occipito-parietal sources. On average, the location of the vMMN sources was 2 cm away from the sources for equiprobable stimulus localized in the same time window. Auditory MEG studies also showed separate generators for the N1 and MMN (Leva¨nen et al. 1996). Our spatio-temporal source localization study supports this finding in the visual modality. Our results are also consistent with the previous EEG and MEG studies that reported deviance-related activity in the middle occipital gyrus and cuneus (Kimura et al. 2010; Urakawa et al. 2010). However, neither of the two previous studies on the localization of the vMMN generators controlled for attention, as the tasks were the detection of the deviant stimuli (Kimura et al. 2010) or watching a silent video (Urakawa et al. 2010). In the present study, we used a task which was designed to control better for attentional effects (Heslenfeld and Kurpershoek 2006). Finally, vMMN responses for gratings with different SF’s depended on the physical features of the stimulus, probably reflecting differential processing of high SF information by parvocellular and low SF information by magnocellular visual channels (Tapia and Breitmeyer 2011). Auditory MMN was also reported to be stimulus dependent (Na¨a¨ta¨nen 1992), and Sulykos and Czigler (2011) reported a feature-related vMMN. Our data provide new evidence that stimulus-dependent mechanisms are underlying the vMMN as well, indicating that the vMMN is not caused by a generic mechanism sensitive to regularity violation as such.
123
Acknowledgments This study was supported by the European Union (COST B27 Action) and Croatian Ministry of Science, Education, and Sport (Grant 119-1081870-1252).
References Astikainen P, Lillstrang E, Ruusuvirta T (2008) Visual mismatch negativity for changes in orientation—sensory memory-dependent response. Eur J Neurosci 28:2319–2324 Czigler I, Balazs Winkler I (2002) Memory-based detection of taskirrelevant visual changes. Psychophysiology 39:869–873 Garrido MI, Kilner J, Stephan KE, Friston KJ (2009) The mismatch negativity: a review of underlying mechanisms. Clin Neurophysiol 120:453–463 Heslenfeld DJ (2003) Visual mismatch negativity. In: Polich J (ed) Detection of change: event-related potential and fMRI findings. Kluwer Academic Publishers, Boston, pp 41–59 Heslenfeld DJ, Kurpershoek CE (2006) Pre-attentive change detection in vision. 12th Annual Meeting of the Organization for Human Brain Mapping, Neuroimage Special Issue S30 Josef Golubic S, Susac A, Grilj V, Ranken D, Huonker R, Haueisen J, Supek S (2011) Size matters: MEG empirical and simulation study on source localization of the earliest visual activity in the occipital cortex. Med Biol Eng Comput 49:545–554 Kimura M (2012) Visual mismatch negativity and unintentional temporal-context-based prediction in vision. Int J Psychophysiol 83:144–155 Kimura M, Katayama J, Ohira H, Schro¨ger E (2009) Visual mismatch negativity: new evidence from the equiprobable paradigm. Psychophysiology 46:402–409 Kimura M, Ohira H, Schro¨ger E et al (2010) Localizing sensory and cognitive systems for pre-attentive visual deviance detection: an sLORETA analysis of the data of Kimura et al. (2009). Neurosci Lett 485:198–203 Leva¨nen S, Ahonen A, Hari R, McEvoy L, Sams M (1996) Deviant auditory stimuli activate human left and right auditory cortex differently. Cereb Cortex 6:288–296 Na¨a¨ta¨nen R (1992) Attention and brain function. Lawrence Erlbaum Associates, Hillsdale Na¨a¨ta¨nen R, Gaillard AWK, Ma¨ntysalo S (1978) Early selectiveattention effect on evoked potential reinterpreted. Acta Psychol 42:313–329 Schro¨ger E, Wolff C (1996) Mismatch response of the human brain to changes in sound location. Neuroreport 7:3005–3008 Sulykos I, Czigler I (2011) One plus one is less than two: visual features elicit non-additive mismatch-related brain activity. Brain Res 1398:64–71 Tapia E, Breitmeyer BG (2011) Visual consciousness revisited: magnocellular and parvocellular contributions to conscious and nonconscious vision. Psychol Sci 22:934–942 Urakawa T, Inui K, Yamashiro K, Tanaka E, Kakigi R (2010) Cortical dynamics of visual change detection based on sensory memory. Neuroimage 52:302–308 Winkler I, Czigler I (2012) Evidence from auditory and visual eventrelated potential (ERP) studies of deviance detection (MMN and vMMN) linking predictive coding theories and perceptual object representations. Int J Psychophysiol 83:132–143
