- Email: [email protected]
no-go performance in patients with schizophrenia
222 were confined to the visual cortex (fusiform and lateral occipital areas). Post-training Control Encode signal patterns were similar to the Task Encode pattern. These data indicate significant visual encode abnormalities in the SZ group before and after extensive training and learning. This is consistent with an "early" defect in perceptual information processing.
THE SPECIFICITY OF CONTEXT PROCESSING DEFICITS ASSOCIATED WITH HYPOFRONTALITY IN SCHIZOPHRENIC PATIENTS: AN EVENT-RELATED FMRI STUDY A. J. H o l m e s , * A. W. M a c D o n a l d , C. S. Carter, D. M. Barch, A. V. Stenger, J. D. C o h e n
Psychiatry, University of Pittsburgh, Pittsburgh, PA, USA Schizophrenic patients frequently demonstrate hypofrontality in tasks that require executive processing; however the questions still remains as to whether or not the observed prefrontal cortex dysfunctions are specific to schizophrenia. Context processing is conceptualized to be the executive function associated with attention and working memory processes. Impairment in schizophrenic patients' ability to represent and maintain context information has been previously reported in a number of studies. To examine the question of the specificity of a context processing deficit to schizophrenia, we administered a version of the expectancy AX task designed to assess context processing in a group of healthy controls (n=9), depressed patient controls (n=l 0), and patients with schizophrenia (n=8) while inside a 1.5T signa GE whole body scanner. In this task, 70% of the trials were AX sequences, with the rest AY, BX, and BY sequences (B is the non-A cue, Y represents any non-X probe). BX false alarms indicate poor attention and working memory. AY false alarms served as a general difficulty control. The behavioral performance was consistent with a context processing deficit in patients with schizophrenia, but not those with depression who showed a trend toward increased AY errors. The imaging data showed patterns of decreased activations in the right medial frontal g y m s for the schizophrenic patients while representing the context provided by the cue. Schizophrenic patients showed a dysfunction while maintaining the cue information in the medial frontal gyms, as well as a dysfunction when preparing to overcome a pre-potent response over time in BA10. In contrast, depressed patients showed bilateral decreases in the activation in the superior and medial frontal gyri when required to maintain the cue. This study demonstrates that it is possible to examine the specificity of the context processing impairment and cortical dysfunction in schizophrenia. Evaluating the specificity of cognitive impairments and regional brain dysfunction in schizophrenia provides important clues as to the neural basis of schizophrenia symptoms.
NEURAL MECHANISM OF PREDICTIVE SMOOTH PURSUIT ABNORMALITIES IN SCHIZOPHRENIA L. E. H o n g , * M. T a g a m e t s , M. Avila, H. H o l c o m b , G. K. T h a k e r
Department of Psychiatry, University of Maryland, Maryland Psychiatric Research Center, Baltimore, MD, USA Smooth pursuit eye movement (SPEM) abnormalities are observed in about 40% schizophrenic patients. However, the precise neural mechanisms underlying the abnormalities are not known. SPEM is
14. Neuroimaging, Functional maintained by a combination of response to the motion of target image on the retina (retinal motion), and a predictive response based on an internal representation of target motion in visual space. This can be mathematically expressed as the following: Maintenance pursuit = Response to retinal motion signal + Predictive response. Predictive pursuit is found abnormal in both schizophrenic patients and their relatives. The purpose of this study is to investigate the neural mechanism of abnormal predictive smooth pursuit in schizophrenic patients using functional MRI. Twelve schizophrenic patients and 12 normal controls were matched on performance during the maintenance of pursuit (by matching on closed-loop pursuit gain, a standard measure of pursuit maintenance). Previously we have noted that even when schizophrenia spectrum subjects show normal closedloop gain, they still show significantly reduced predictive pursuit gain. These matched subjects underwent functional MRI while pursuing a sinusoidal, smoothly moving target. Multislice EPI in a Philips 1.5 T magnet was used to produce 28 contiguous, 5-ram thick axial slices covering the entire brain in 4 blocks of the same task for each subject. The test and re-test results for maintenance pursuit gain and predictive pursuit gain were stable when tested 1 month or more apart. Compared to normal controls, schizophrenic patients had reduced activation at supramarginal gyms, superior temporal gyms, anterior/posterior cerebellum, cingulate, and frontal eye field. Schizophrenic patients had increased activation in area 18 and area 19 (V5, or MT). The results suggest that when schizophrenic and healthy control subjects were well-matched on performance, the two groups used different neural circuits to maintain pursuit. Healthy subjects showed increased activation of the neural pathway implicated in predictive component, while patients showed increased activation of posterior pathway in processing of retinal motion signals.
FUNCTIONAL MRI AT 3T DURING A GO/NOGO PERFORMANCE IN PATIENTS WITH SCHIZOPHRENIA M. J. H o p t m a n , * K. A. Nolan, E. M. Weiss, R. M. Bilder, E S. O ' D o n n e l l , D. N. Guilfoyle, E. Z a r a h n , J. Volavka
Clinical Research, Nathan Kline lnstituteJbr Psychiatric Research, Orangeburg, NY, USA Impulsivity is a significant component of the behavioral and neurocognitive abnormalities seen in patients with schizophrenia. It is associated with high levels of substance abuse, increased risk for aggression, and suicide. The Go/No-Go task has been used extensively to study response inhibition in a variety of populations. It requires the subject to make a response in the presence of the "go" stimulus and to make no response in the presence of the "no-go" stimulus. Commission errors (responding in the presence of the nogo stimulus) are interpreted as failures of response inhibition, and are increased in patients with frontal lesions, in aggressive male adolescents, and in psychopaths. Fifteen inpatients with schizophrenia or schizoaffective disorder were tested on a Go/No-Go task in a functional magnetic resonance imaging study. Scans were acquired using a gradient echo Echo Planar Imaging sequence (TR = 3000 ms, TE = 40 ms, 64 x 64 matrix, FOV = 22.4 cm, slice thickness = 3.5 mm, 3i slices, no gap) on a 3T scanner. Violent behavior was assessed with the Life History of Aggression, and patients were divided into more violent and less violent groups using a median split. The scans were processed and analyzed using SPM99. A p value of.001 and an extent threshold of 10 contiguous voxels was set for all comparisons. Random effects analyses on the contrast of no-go vs. rest, with violence group as a covariate, revealed that the more violent patients showed greater activation in the right anterior cingulate (Brodmann
International Congress on Schizophrenia Research 2003
14. Neuroimaging, Functional
223
Area [BA] 32), the left superior frontal gyms (BA8), the left medial frontal gyms (BA46), the left superior temporal gyms (BA22), the right inferior frontal gyms (BA45), and the thalamus. The current results suggest that more violent patients activate a fronto-temporolimbic circuit in situations requiring response inhibition. The implications of these results for impulsivity in schizophrenia will be discussed, including possible neural inefficiency in fronto-temporo-limbic circuits in more violent patients.
AUDITORY RESPONSE SHOWING BRAIN SOURCE ABNORMALITIES IN SCHIZOPHRENIA M. H u a n g , * J. C. Edgar, R. J. T h o m a , R M. Hanlon, S. N. M o s e s , R. R. Lee, K. M. Paulson, M. R W e i s e n d , J. R. Bustillo, L. E. Adler, G. A. Miller, J. M. C a n i v e
Center for Functional Brain Imaging, New Mexico VA Health Care System, Albuquerque, NM, USA Abnormalities in early cortical responses to auditory stimuli have been described using EEG and MEG in schizophrenia patients. MEG studies identify the superior temporal gyms (STG) as the region predominantly contributing to the early auditory magnetic response. However, the specific brain generators contributing to this response and to its abnormality in schizophrenia have not been as well characterized. In the present study, using an integrated approach, the contribution of the STG generators to the EEG was examined. Auditory responses were recorded simultaneously using EEG and MEG from 20 patients with schizophrenia and 19 control subjects. Bilateral STG sources and their time courses were obtained using MEG for the 30-100 ms post-stimulus interval. The MEG STG source time courses were then used to predict the EEG signal at electrode Cz for the same time period. In control subjects, the bilateral MEG STG time courses predicted the variance in Cz EEG very well, with percent-variance-explained (PVE) of 97.1 +/- 2.1% (mean and SD). In patients with schizophrenia, bilateral STG time courses accounted for less variance, PVE 86.7 +/- 9.9%. The group difference in PVE was reliable (p<.0002). In addition, there was little overlap between the patient and control samples in range of PVE values. After STG activity was removed from Cz EEG, the residual signal was dominated by a 40 Hz signal. This indicates that the remaining variance in EEG is probably contributed by additional brain generators not in STG. Thus, present results provide evidence of specific STG sources that are less active in schizophrenia and also suggest that additional source(s) are active, or more active, in schizophrenia relative to nonpatients.
TEMPORAL CHARACTERISTICS OF HAEMODYNAMIC RESPONSE IN PLANUM TEMPORALE MODULATED BY LATERALITY OF HALLUCINATION-LIKESPEECH M. D. Hunter,* R Tandon, I. D. Wilkinson, S. A. Spence, R W. W o o d r u f f
SCANLab, Department of Psychiatry, University of Sheffield, Sheffield, South Yorkshire, United Kingdom We aimed, using functional magnetic resonance imaging (fMRI), to investigate the effect of stimulus laterality on temporal characteristics of the haemodynamic response to hallucination-like speech, in the planum temporale (PT). 5 healthy, right-handed, males were stud-
ied. Over electrostatic headphones, hallucination-like speech stimuli were either central o1"lateralised to one side or the other by adjustment of the inter~aural amplitude ratio, fMRI was performed on a 1.5T system at Sheffield University (450 time points; repetition time= 1s). In an event-related design, stimuli were presented at pseudo-random intervals. Subjects underwent 3 functional scans; the spatial location of every individual stimulus was counterbalanced between scans. Images were analysed using statistical parametric mapping in SPM99. Group analysis of response to all speech was thresholded at p<0.05, corrected, and demonstrated activation in left and right superior temporal gyri. We then analysed the time course of regions of interest derived from spheres of radius 10mm centred on the probabilistic maxima of the left and right PT (Westbury et al., Cereb Cortex 1999; 9: 392-405). In the left PT, ipsilateral speech was associated with mean latency to peak blood oxygenation level dependent (BOLD) response of 6.41s but contralateral speech was associated with significantly longer mean latency to peak BOLD response of 7.26s (Wilcoxon signed-ranks test; Z=1.89; 1-tailed p=0.03). Contralateral speech was also associated with mean peak BOLD response of l 0.34% greater amplitude than ipsilateral speech (Z=1.69; 1-tailed p=0.04). In the right PT, there was no difference between ipsilateral and contralateral speech in mean latency to peak BOLD response (6.66s & 6.61s, respectively). Contralateral speech was associated with mean peak BOLD response of 32.00% greater amplitude than ipsilateral speech (Z=3.09; 1-tailed p=0.001). In left and right PT, central speech was associated with a BOLD response curve that was intermediate between ipsilateral and contralateral speech. Our findings support the idea of a more sustained neural response to contralateral versus ipsilateral speech, specific to the left PT. This phenomenon might reflect functional specialisation for language processing in the left PT. Our paradigm may be a useful probe of lateralised brain organisation in patients with actual hallucinations.
IMPACT OF NICOTINE ON BRAIN FUNCTION IN SMOKERS WITH SCHIZOPHRENIA L. K. Jacobsen,* W. E. M e n c l , D. C. D ' S o u z a , K. R. Pugh, R Skudlarski, W. Abi-Saab, J. H. Krystal
Psychiat~% VA Connecticut Healthcare System, WesttIaven, CT, USA The prevalence of cigarette smoking among patients with schizoprhenia has been reported to be 40% to 100% higher than among patients with other psychiatric disorders and up to three times higher than the prevalence in the general population. Evidence suggests that smokers with schizphrenia smoke more heavily and extract more nicotine from smoking than do smokers with or without other psychiatric disorders. Consistent with this, rates of sustained abstinence are extremely low in this population. Nicotine, the reinforcing and addicting component of tobacco smoke, has complex effects on cognitive processing. In the absence of nicotine withdrawal, nicotine can improve attention and motor speed and, under some conditions, has been shown to improve memory. However, nicotine withdrawal is associated with significant deficits in both memory and attention. In smokers with schizophrenia, nicotine has been shown to improve eye movement abnormalities and spatial working memory, and transiently reverse the auditory P50 gating deficit associated with this disorder. In this study, the effect of nicotine and of nicotine withdrawal on activation of neural circuits that mediate performance of tasks involving verbal working memory and selective attention is examined in smokers with and without schizophrenia using fMRI.
International Congress on Schizophrenia Research 2003
THE SPECIFICITY OF CONTEXT PROCESSING DEFICITS ASSOCIATED WITH HYPOFRONTALITY IN SCHIZOPHRENIC PATIENTS: AN EVENT-RELATED FMRI STUDY A. J. H o l m e s , * A. W. M a c D o n a l d , C. S. Carter, D. M. Barch, A. V. Stenger, J. D. C o h e n
Psychiatry, University of Pittsburgh, Pittsburgh, PA, USA Schizophrenic patients frequently demonstrate hypofrontality in tasks that require executive processing; however the questions still remains as to whether or not the observed prefrontal cortex dysfunctions are specific to schizophrenia. Context processing is conceptualized to be the executive function associated with attention and working memory processes. Impairment in schizophrenic patients' ability to represent and maintain context information has been previously reported in a number of studies. To examine the question of the specificity of a context processing deficit to schizophrenia, we administered a version of the expectancy AX task designed to assess context processing in a group of healthy controls (n=9), depressed patient controls (n=l 0), and patients with schizophrenia (n=8) while inside a 1.5T signa GE whole body scanner. In this task, 70% of the trials were AX sequences, with the rest AY, BX, and BY sequences (B is the non-A cue, Y represents any non-X probe). BX false alarms indicate poor attention and working memory. AY false alarms served as a general difficulty control. The behavioral performance was consistent with a context processing deficit in patients with schizophrenia, but not those with depression who showed a trend toward increased AY errors. The imaging data showed patterns of decreased activations in the right medial frontal g y m s for the schizophrenic patients while representing the context provided by the cue. Schizophrenic patients showed a dysfunction while maintaining the cue information in the medial frontal gyms, as well as a dysfunction when preparing to overcome a pre-potent response over time in BA10. In contrast, depressed patients showed bilateral decreases in the activation in the superior and medial frontal gyri when required to maintain the cue. This study demonstrates that it is possible to examine the specificity of the context processing impairment and cortical dysfunction in schizophrenia. Evaluating the specificity of cognitive impairments and regional brain dysfunction in schizophrenia provides important clues as to the neural basis of schizophrenia symptoms.
NEURAL MECHANISM OF PREDICTIVE SMOOTH PURSUIT ABNORMALITIES IN SCHIZOPHRENIA L. E. H o n g , * M. T a g a m e t s , M. Avila, H. H o l c o m b , G. K. T h a k e r
Department of Psychiatry, University of Maryland, Maryland Psychiatric Research Center, Baltimore, MD, USA Smooth pursuit eye movement (SPEM) abnormalities are observed in about 40% schizophrenic patients. However, the precise neural mechanisms underlying the abnormalities are not known. SPEM is
14. Neuroimaging, Functional maintained by a combination of response to the motion of target image on the retina (retinal motion), and a predictive response based on an internal representation of target motion in visual space. This can be mathematically expressed as the following: Maintenance pursuit = Response to retinal motion signal + Predictive response. Predictive pursuit is found abnormal in both schizophrenic patients and their relatives. The purpose of this study is to investigate the neural mechanism of abnormal predictive smooth pursuit in schizophrenic patients using functional MRI. Twelve schizophrenic patients and 12 normal controls were matched on performance during the maintenance of pursuit (by matching on closed-loop pursuit gain, a standard measure of pursuit maintenance). Previously we have noted that even when schizophrenia spectrum subjects show normal closedloop gain, they still show significantly reduced predictive pursuit gain. These matched subjects underwent functional MRI while pursuing a sinusoidal, smoothly moving target. Multislice EPI in a Philips 1.5 T magnet was used to produce 28 contiguous, 5-ram thick axial slices covering the entire brain in 4 blocks of the same task for each subject. The test and re-test results for maintenance pursuit gain and predictive pursuit gain were stable when tested 1 month or more apart. Compared to normal controls, schizophrenic patients had reduced activation at supramarginal gyms, superior temporal gyms, anterior/posterior cerebellum, cingulate, and frontal eye field. Schizophrenic patients had increased activation in area 18 and area 19 (V5, or MT). The results suggest that when schizophrenic and healthy control subjects were well-matched on performance, the two groups used different neural circuits to maintain pursuit. Healthy subjects showed increased activation of the neural pathway implicated in predictive component, while patients showed increased activation of posterior pathway in processing of retinal motion signals.
FUNCTIONAL MRI AT 3T DURING A GO/NOGO PERFORMANCE IN PATIENTS WITH SCHIZOPHRENIA M. J. H o p t m a n , * K. A. Nolan, E. M. Weiss, R. M. Bilder, E S. O ' D o n n e l l , D. N. Guilfoyle, E. Z a r a h n , J. Volavka
Clinical Research, Nathan Kline lnstituteJbr Psychiatric Research, Orangeburg, NY, USA Impulsivity is a significant component of the behavioral and neurocognitive abnormalities seen in patients with schizophrenia. It is associated with high levels of substance abuse, increased risk for aggression, and suicide. The Go/No-Go task has been used extensively to study response inhibition in a variety of populations. It requires the subject to make a response in the presence of the "go" stimulus and to make no response in the presence of the "no-go" stimulus. Commission errors (responding in the presence of the nogo stimulus) are interpreted as failures of response inhibition, and are increased in patients with frontal lesions, in aggressive male adolescents, and in psychopaths. Fifteen inpatients with schizophrenia or schizoaffective disorder were tested on a Go/No-Go task in a functional magnetic resonance imaging study. Scans were acquired using a gradient echo Echo Planar Imaging sequence (TR = 3000 ms, TE = 40 ms, 64 x 64 matrix, FOV = 22.4 cm, slice thickness = 3.5 mm, 3i slices, no gap) on a 3T scanner. Violent behavior was assessed with the Life History of Aggression, and patients were divided into more violent and less violent groups using a median split. The scans were processed and analyzed using SPM99. A p value of.001 and an extent threshold of 10 contiguous voxels was set for all comparisons. Random effects analyses on the contrast of no-go vs. rest, with violence group as a covariate, revealed that the more violent patients showed greater activation in the right anterior cingulate (Brodmann
International Congress on Schizophrenia Research 2003
14. Neuroimaging, Functional
223
Area [BA] 32), the left superior frontal gyms (BA8), the left medial frontal gyms (BA46), the left superior temporal gyms (BA22), the right inferior frontal gyms (BA45), and the thalamus. The current results suggest that more violent patients activate a fronto-temporolimbic circuit in situations requiring response inhibition. The implications of these results for impulsivity in schizophrenia will be discussed, including possible neural inefficiency in fronto-temporo-limbic circuits in more violent patients.
AUDITORY RESPONSE SHOWING BRAIN SOURCE ABNORMALITIES IN SCHIZOPHRENIA M. H u a n g , * J. C. Edgar, R. J. T h o m a , R M. Hanlon, S. N. M o s e s , R. R. Lee, K. M. Paulson, M. R W e i s e n d , J. R. Bustillo, L. E. Adler, G. A. Miller, J. M. C a n i v e
Center for Functional Brain Imaging, New Mexico VA Health Care System, Albuquerque, NM, USA Abnormalities in early cortical responses to auditory stimuli have been described using EEG and MEG in schizophrenia patients. MEG studies identify the superior temporal gyms (STG) as the region predominantly contributing to the early auditory magnetic response. However, the specific brain generators contributing to this response and to its abnormality in schizophrenia have not been as well characterized. In the present study, using an integrated approach, the contribution of the STG generators to the EEG was examined. Auditory responses were recorded simultaneously using EEG and MEG from 20 patients with schizophrenia and 19 control subjects. Bilateral STG sources and their time courses were obtained using MEG for the 30-100 ms post-stimulus interval. The MEG STG source time courses were then used to predict the EEG signal at electrode Cz for the same time period. In control subjects, the bilateral MEG STG time courses predicted the variance in Cz EEG very well, with percent-variance-explained (PVE) of 97.1 +/- 2.1% (mean and SD). In patients with schizophrenia, bilateral STG time courses accounted for less variance, PVE 86.7 +/- 9.9%. The group difference in PVE was reliable (p<.0002). In addition, there was little overlap between the patient and control samples in range of PVE values. After STG activity was removed from Cz EEG, the residual signal was dominated by a 40 Hz signal. This indicates that the remaining variance in EEG is probably contributed by additional brain generators not in STG. Thus, present results provide evidence of specific STG sources that are less active in schizophrenia and also suggest that additional source(s) are active, or more active, in schizophrenia relative to nonpatients.
TEMPORAL CHARACTERISTICS OF HAEMODYNAMIC RESPONSE IN PLANUM TEMPORALE MODULATED BY LATERALITY OF HALLUCINATION-LIKESPEECH M. D. Hunter,* R Tandon, I. D. Wilkinson, S. A. Spence, R W. W o o d r u f f
SCANLab, Department of Psychiatry, University of Sheffield, Sheffield, South Yorkshire, United Kingdom We aimed, using functional magnetic resonance imaging (fMRI), to investigate the effect of stimulus laterality on temporal characteristics of the haemodynamic response to hallucination-like speech, in the planum temporale (PT). 5 healthy, right-handed, males were stud-
ied. Over electrostatic headphones, hallucination-like speech stimuli were either central o1"lateralised to one side or the other by adjustment of the inter~aural amplitude ratio, fMRI was performed on a 1.5T system at Sheffield University (450 time points; repetition time= 1s). In an event-related design, stimuli were presented at pseudo-random intervals. Subjects underwent 3 functional scans; the spatial location of every individual stimulus was counterbalanced between scans. Images were analysed using statistical parametric mapping in SPM99. Group analysis of response to all speech was thresholded at p<0.05, corrected, and demonstrated activation in left and right superior temporal gyri. We then analysed the time course of regions of interest derived from spheres of radius 10mm centred on the probabilistic maxima of the left and right PT (Westbury et al., Cereb Cortex 1999; 9: 392-405). In the left PT, ipsilateral speech was associated with mean latency to peak blood oxygenation level dependent (BOLD) response of 6.41s but contralateral speech was associated with significantly longer mean latency to peak BOLD response of 7.26s (Wilcoxon signed-ranks test; Z=1.89; 1-tailed p=0.03). Contralateral speech was also associated with mean peak BOLD response of l 0.34% greater amplitude than ipsilateral speech (Z=1.69; 1-tailed p=0.04). In the right PT, there was no difference between ipsilateral and contralateral speech in mean latency to peak BOLD response (6.66s & 6.61s, respectively). Contralateral speech was associated with mean peak BOLD response of 32.00% greater amplitude than ipsilateral speech (Z=3.09; 1-tailed p=0.001). In left and right PT, central speech was associated with a BOLD response curve that was intermediate between ipsilateral and contralateral speech. Our findings support the idea of a more sustained neural response to contralateral versus ipsilateral speech, specific to the left PT. This phenomenon might reflect functional specialisation for language processing in the left PT. Our paradigm may be a useful probe of lateralised brain organisation in patients with actual hallucinations.
IMPACT OF NICOTINE ON BRAIN FUNCTION IN SMOKERS WITH SCHIZOPHRENIA L. K. Jacobsen,* W. E. M e n c l , D. C. D ' S o u z a , K. R. Pugh, R Skudlarski, W. Abi-Saab, J. H. Krystal
Psychiat~% VA Connecticut Healthcare System, WesttIaven, CT, USA The prevalence of cigarette smoking among patients with schizoprhenia has been reported to be 40% to 100% higher than among patients with other psychiatric disorders and up to three times higher than the prevalence in the general population. Evidence suggests that smokers with schizphrenia smoke more heavily and extract more nicotine from smoking than do smokers with or without other psychiatric disorders. Consistent with this, rates of sustained abstinence are extremely low in this population. Nicotine, the reinforcing and addicting component of tobacco smoke, has complex effects on cognitive processing. In the absence of nicotine withdrawal, nicotine can improve attention and motor speed and, under some conditions, has been shown to improve memory. However, nicotine withdrawal is associated with significant deficits in both memory and attention. In smokers with schizophrenia, nicotine has been shown to improve eye movement abnormalities and spatial working memory, and transiently reverse the auditory P50 gating deficit associated with this disorder. In this study, the effect of nicotine and of nicotine withdrawal on activation of neural circuits that mediate performance of tasks involving verbal working memory and selective attention is examined in smokers with and without schizophrenia using fMRI.
International Congress on Schizophrenia Research 2003