Comment by Kevin M. Spencer and Robert W. McCarley
Uhlhaas and colleagues present very interesting findings in their study examining high-frequency inter-areal synchronization during Gestalt perception in schizophrenia. They used an elegant task in which the responses to Gestalt stimuli (“Mooney” faces) were compared to the same stimuli that were inverted or scrambled as a control condition. The motivation for using this kind of task is that it should tap the feature-binding processes that have been proposed to be mediated by high-frequency oscillatory synchronization (e.g., Singer, 1999). The major findings of the study were that in control subjects, Gestalt perception was associated with periods of phase synchronization between electrode sites in the beta (15-30 Hz) frequency band. Schizophrenia patients also showed phase synchrony in the beta band during face perception, but it was reduced in magnitude and delayed in latency with respect to the controls. For schizophrenics, the degree of phase synchrony in the Gestalt condition was positively correlated with positive symptoms on the PANSS, particularly delusions and hallucinations. In comparing the findings of Uhlhaas et al. to our work and other published data, we note some interesting aspects of their study.
At the most basic level, this study adds to the growing literature demonstrating deficits in measures of high-frequency oscillatory synchrony in the EEG in schizophrenia. Previously we reported similar analyses of phase synchrony (also termed “phase coherence”) in a comparable Gestalt perception task employing Kanisza squares as stimuli (Spencer et al., 2003). (Our task was inspired by that of Rodriguez et al. , after which the present study was patterned.) While we found that schizophrenia patients showed phase synchrony abnormalities around 40 Hz, Uhlhaas et al. found abnormalities in the beta range. The difference in frequency is noteworthy. One possible explanation is that Mooney faces are more complex and/or difficult to perceive than Kanisza squares and thus involve the synchronization of larger numbers of brain regions. (The error rates in the Uhlhaas et al. study were higher than in our study.) Some researchers have proposed that synchrony between distant brain regions is more likely to occur at lower frequencies due to axonal conduction delays. However, Rodriguez et al. found phase synchrony predominantly at ~40 Hz with Mooney face stimuli, so it is not clear why Uhlhaas et al. observed synchrony in the beta rather than the gamma band.
Another interesting aspect of the study was that for schizophrenics, phase synchrony in the Gestalt condition was positively correlated with positive symptoms. It is quite common to find that some neurophysiological or neuroanatomical measure is negatively correlated with symptoms in neuropsychiatric disorders, so positive correlations are rare. We found a conceptually similar pattern of positive correlations between positive symptoms and another measure of oscillatory synchrony in Gestalt perception (Spencer et al., 2004). Furthermore, Lee et al. (2003b) have also reported a similar pattern of correlations with a different measure of synchrony in an auditory oddball task. Thus, the findings from several groups may be pointing toward a more general association between positive symptoms and increased phase synchrony. We recently proposed that such a relationship might be related to increased excitability in sensory and association cortex (Spencer and McCarley, 2005).
One potential confound that Uhlhaas et al. noted was that the schizophrenic abnormalities they observed might be attributable to a deficit in face perception (e.g., Onitsuka et al., 2006), rather than Gestalt perception. While we are not convinced that simply excluding particular electrodes from analysis could alleviate this confound, we note that the “confound” could actually go in the opposite direction: abnormalities in face perception in schizophrenia might be related to a more general deficit in Gestalt perception. This hypothesis deserves further examination.
It should be noted that the inter-electrode phase synchrony measure utilized by Uhlhaas et al. and ourselves (Spencer et al., 2003), the similar measure used by the Gordon group (Lee et al., 2003a; 2003b), and more traditional measures of power coherence are not without at least two potential confounds: the influence of the reference electrode and volume conduction. First, a change in synchrony or coherence between two electrodes could be caused by activity measured at the reference electrode(s) to which the electrodes of interest are referenced. Second, volume conduction can introduce spurious synchrony or coherence between two electrodes that both pick up the same activity. The pitfalls of these measures have been analyzed in detail (e.g., Guevara et al., 2005; Nunez et al., 1997) and should be taken seriously. How these potential confounds might influence the results of the above-cited studies needs to be determined.
Lastly, we would like to clear up a couple of inaccuracies about our work that were stated in the paper. First, neither of our studies examined induced power, but rather phase locking. Second, the patients in our studies did show statistically significant deficits in task performance, although they were not as severe as in the Uhlhaas et al. study.
To conclude, the study of Uhlhaas et al. is a timely and welcome addition to the literature on neural dynamics in schizophrenia. We are encouraged that the points of convergence between this and other published studies indicate that researchers are heading in the right direction toward understanding neural dysfunction in schizophrenia.
Guevara R, Velazquez JL, Nenadovic V, Wennberg R, Senjanovic G, Dominguez LG. Phase synchronization measurements using electroencephalographic recordings: what can we really say about neuronal synchrony?
Nunez PL, Srinivasan R, Westdorp AF, Wijesinghe RS, Tucker DM, Silberstein RB, Cadusch PJ. EEG coherency. I: Statistics, reference electrode, volume conduction, Laplacians, cortical imaging, and interpretation at multiple scales.
Electroencephalogr Clin Neurophysiol. 1997 Nov;103(5):499-515. Review.
Lee KH, Williams LM, Breakspear M, Gordon E. Synchronous gamma activity: a review and contribution to an integrative neuroscience model of schizophrenia.
Brain Res Brain Res Rev. 2003 Jan;41(1):57-78. Review.
Lee KH, Williams LM, Haig A, Gordon E. "Gamma (40 Hz) phase synchronicity" and symptom dimensions in schizophrenia.
Cognit Neuropsychiatry. 2003 Feb;8(1):57-71.
Onitsuka T, Niznikiewicz MA, Spencer KM, Frumin M, Kuroki N, Lucia LC, Shenton ME, McCarley RW. Functional and structural deficits in brain regions subserving face perception in schizophrenia.
Am J Psychiatry. 2006 Mar;163(3):455-62.
Rodriguez E, George N, Lachaux JP, Martinerie J, Renault B, Varela FJ. Perception's shadow: long-distance synchronization of human brain activity.
Nature. 1999 Feb 4;397(6718):430-3.
Spencer KM, McCarley RW (2005). Visual hallucinations, attention, and neural circuitry: perspectives from schizophrenia research. Behav Brain Sci 28:774.
Spencer KM, Nestor PG, Niznikiewicz MA, Salisbury DF, Shenton ME, McCarley RW. Abnormal neural synchrony in schizophrenia.
J Neurosci. 2003 Aug 13;23(19):7407-11.
Spencer KM, Nestor PG, Perlmutter R, Niznikiewicz MA, Klump MC, Frumin M, Shenton ME, McCarley RW. Neural synchrony indexes disordered perception and cognition in schizophrenia.
Proc Natl Acad Sci U S A. 2004 Dec 7;101(49):17288-93. Epub 2004 Nov 16.
View all comments by Kevin Spencer