Supplemental Information
Supplemental Methods
Screening
Subjects were assessed using a structured interview for diagnosing psychiatric disorder, SCID-I/NP (First et al, 2002). Psychotic-like symptoms were assessed with the Perceptual-Aberration – Magical Ideation and the Revised Social Anhedonia Scale (Chapman et al, 1978; Eckblad et al, unpublished; Eckblad et al, 1983). First-degree family history of psychiatric disorder was assessed with the Family History Assessment Module (Janca et al, 1997). An outside informant (typically a spouse, parent, or close relative) was asked to verify psychiatric, substance abuse, and medical history. Participants received a physical exam including urine drug screen, breathalyzer test, full blood count, and clinical chemistry. They were eliminated if they were not in good physical health, had an Axis I psychiatric disorder or a first-degree relative with an Axis I psychiatric disorder, or scored higher than 2 standard deviations above the norm on the tests of psychotic-like symptoms. Three smokers were included. Ketamine-related percent signal change alterations during the encoding and early maintenance phase (EEM) were not related to age, gender, or average plasma ketamine level obtained, and these variables were not further analyzed.
Scan Parameters
We acquired high-resolutionT1-weighted structural images with a 3D magnetization prepared gradient echo (MPRAGE) sequence (160 contiguous slices, slice thickness = 1mm slices, matrix = 2562, in-plane resolution = .78 mm2). In addition, a second series of axial-oblique structural images were acquired parallel to the anterior-posterior commissure line (24 contiguous slices, slice thickness = 5 mm, matrix = 2562, in-plane resolution= .78mm). FMRI images were acquired co-planar to this second series of structural images using a T2*-sensitive gradient-recalled single shot echo-planar pulse sequence (24 contiguous slices, slice thickness = 5mm, TR = 1500ms, TE = 30ms, flip angle α = 80, matrix = 642, in-plane resolution = 3.1mm2). Each subject completed 18 scans of 166 image volumes, including six initial images that were later discarded.
Software for Image Analysis
Preprocessing and primary analyses were conducted with the FMRIB Software Library (FSL, http://www.fmrib.ox.ac.uk/fsl). Motion parameters were converted to distance metrics using software written by Syam Gadde, Ph.D., of the Brain Imaging and Analysis Center at Duke University, N.C. To analyze time course activation, we used software tools from the Biomedical Informatics Research Network, an NIH/NHCRR consortium of university imaging centers http://www.nitrc.org/projects/bxh_xcede_tools/). The maps correlating performance and PANSS symptoms with connectivity alterations were computed with custom software written for this purpose.
Pre-processing
Pre-processing and primary analyses were conducted with the FMRIB Software Library (FSL, http://www.fmrib.ox.ac.uk/fsl). Additional software used is described in SI. Data were converted into NIFTI (Neuroimaging Informatics Technology Initiative) format, and brain extraction was performed to delete non-brain elements (Jenkinson et al, 2002). Then images were corrected for slice-timing, smoothed with a Gaussian kernel of FWHM 5mm and normalized with corrected to the middle image of the run for motion using pass at 0.01 Hz. For the analyses of time course, voxel-wise temporal auto-correlation was estimated filtered MCFLIRT(Jenkinson et al, 2002), a single multiplicative factor. Images were high-and corrected using FMRIB’s Improved Linear Model (FILM) (Woolrich et al, 2001).
For all scans, registration was achieved by co-registering each individual’s co-planar image to their high-resolution structural scan. The high-resolution structural scan was then co-registered to the MNI standard brain supplied with FSL. Scans with motion > 2 mm in the z, y or z direction or > 3 degrees of pitch, yaw or roll were eliminated. In the entire sample, a total of two runs were eliminated. Motion correction parameters derived on an individual basis were converted to measures of absolute distance from the center of the brain. These measures were analyzed with the MANOVA approach to repeated measures and no statistically significant difference related to drug, scan, or their interaction was found.
Cluster Correction for Statistical Maps
Procedures for protecting against multiple comparison were derived from Gaussian Random Field Theory and imposed a mean threshold of Z > 1.96 and a corrected significance threshold of p < 0.05 (Forman et al, 1995). This method, implemented in FSL, controlled for family-wise error, based on the chance probability of obtaining a specified number of clusters that exceeded a specified number of voxels above a pre-determined threshold. The statistical maps relating connectivity to performance were computed outside the FSL program, rendering FSL cluster correction unavailable. Therefore, we used 3dClustSim, a cluster correction procedure available in AFNI, to determine a minimum cluster size with Monte Carlo simulation so as to achieve a corrected significance ofp <0.05 with a voxelwise threshold of p < 0.05.
Supplemental Results
DlPFC Connectivity during Rest
Resting connectivity of the dlPFC seed was greater under ketamine than under saline, with and without removal of the global mean signal, Fig. S2. Data were derived from the bolus runs described in the text. They were pre-processed as in the main text, except that data were low-pass filtered with a cut-off frequency of 0.08 Hz. In order to achieve a relatively stable ketamine level, we used the last 50 images of the saline and ketamine bolus scans.
Scan-by-Scan Analyses of Principal Effects
We completed descriptive analyses to evaluate the possibility that effects attributed to ketamine would best be described as fatigue effects. Since these effects would appear as reductions in activation and connectivity over time, we show here scan-by-scan graphs of our key activation and connectivity effects. Our key activation effect was a ketamine-related reduction during the encoding period of our WM task. Fig. S3 shows the average encoding BOLD percent signal change by scan. Inspection of similar graphs by ROI and trial type did not support the hypothesis of a time effect.
Our key connectivity finding was a ketamine-related reduction in R-dlPFC connectivity. To capture this effect, we created a mask representing areas that were significantly reduced by ketamine in our within-subject group comparison map. Then, for every subject, we applied this mask separately to each run and calculated the average z within the mask. This yielded a measure of R-dlPFC connectivity strength for each subject in each run. Finally, we averaged across subjects to obtain a connectivity score for each run, displayed in Fig. S4.
Supplemental Figures
Fig. S4. Percent BOLD signal change by scan during encoding period and four-target task, averaging over ROIsFig. S5. Mean connectivity during task in the R-dlPFC network by scan.
Supplemental Tables
Table S1. Control, Two-Target and Four-Target Task: Mixed-Model Analysis of Activation during Ketamine and Saline Infusion
Effect / Degrees of Freedom / F-Value / pDrug / 1,105 / 3.33 / 0.07
ROI / 2,1024 / 81.54 / <.0001
Drug x ROI / 2,1024 / 1.37 / 0.25
Load / 2,105 / 6.25 / 0.0027
Drug x Load / 2,105 / 0.20 / 0.82
ROI x Load / 4,1024 / 10.39 / <.0001
Phase / 2,1024 / 26.24 / <.0001
Drug x Phase / 2,1024 / 3.40 / 0.03
ROI x Phase / 4,1024 / 26.54 / <.0001
Load x Phase / 4,1024 / 5.73 / 0.0001
ROI x Load x Phase / 8,1024 / 2.34 / 0.01
Drug x ROI X Load / 4, 1024 / 2.67 / 0.03
Table S2. Two-target and Four-target Task Only: Mixed-Model Analysis of Activation During Ketamine and Saline Infusion :
Effect / Degrees of Freedom / F-Value / pDrug / 1,63 / 4.84 / 0.03
ROI / 2,686 / 54.78 / .0001
Drug x ROI / 2,686 / 2.78 / 0.06
Load / 1,63 / 2.28 / 0.13
Drug x Load / 1,63 / 0.07 / 0.78
ROI x Load / 2.686 / 3.86 / 0.02
Phase / 2,686 / 21.08 / .0001
Drug x Phase / 2,686 / 11.17 / .0001
ROI x Phase / 4,686 / 30.27 / .0001
Load x Phase / 2,686 / 12.75 / .0001
Drug x ROI x Load / 2,686 / 2.77 / 0.06
Table S3. Post-hoc Comparisons for Accuracy X Load X ROI
Region-of-Interest / Degrees-of-Freedom / F-value / pInferior Frontal Gyrus / 1, 211 / 4.22 / .04
Middle Frontal Gyrus / 1, 211 / .05 / .83
Superior Frontal Gyrus / 1, 211 / .70 / .40
Note: The reported test statistics and p-values are for the comparisons of slopes for the relationship between accuracy and percent signal change in the 4T and 2T tasks within each region. All slopes except the slope for the 4T task in IFG were positive. None of the slopes were significantly different from zero after correction for multiple comparisons.
Table S4. Clusters Where Change in R-DlPFC Connectivity (Ketamine-Saline) Is Significantly Correlated with Change in Performance (Ketamine-Saline): Descriptive Statistics
Center of MassArea / Cluster # / # of Voxels / Volume (mm3) / X / Y / Z / R
Inferior temporal gyrus, R / 1 / 170 / 1360 / 55 / -26 / -21 / -0.68
Inferior frontal cortex, L / 2 / 229 / 1832 / -44 / 16 / -12 / -0.75
Inferior frontal cortex, R / 3 / 235 / 1880 / 45 / 24 / -11 / -0.74
Bilateral anterior thalamus / 4 / 268 / 2144 / 3 / -8 / 5 / 0.71
Supramarginal gyrus, L / 5 / 273 / 2184 / -48 / -47 / 32 / -0.72
Secondary occipital cortex, R / 6 / 421 / 3368 / 34 / -72 / 11 / -0.73
Table S5. Clusters Where Change in L-DlPFC Connectivity (Ketamine-Saline) Is Significantly Correlated with Change in Performance (Ketamine-Saline): Descriptive Statistics
Center of MassArea / Cluster / Volume / Volume (mm3) / X / Y / Z / r
Middle temporal gyrus, R / 1 / 219 / 1752 / 55 / -12 / -22 / 0.64
Cerebellum, bilateral / 2 / 252 / 2016 / -1 / -52 / -17 / -0.72
Occipital lobe, L / 3 / 266 / 2128 / -23 / -75 / 21 / -0.66
Anterior thalamus/Fornix / 4 / 477 / 3816 / -1 / 2 / -4 / 0.67
References
Chapman LJ, Chapman JP, Raulin ML (1978). Body image aberration in schizophrenia. Journal of Abnormal Psychology 87: 399-407.
Eckblad MJ, Chapman LJ, Chapman JP, Mishlove M (unpublished). The revised social anhedonia scale.
Eckblad ML, Chapman LJ, Chapman JP (1983). Magical ideation as an indicator of schizotypy. Journal of Consulting and Clinical Psychology 51: 215-222.
First M, Spitzer R, Gibbon M, Williams J (2002). Structured Clinical Interview for DSM-IV Axis I Disorders, Research Version, Non-patient Edition Biometrics Research, New York State Psychiatric Institute: New York.
Forman SD, Cohen JD, Fitzgerald M, Eddy WF, Mintun MA, Noll DC (1995). Improved assessment of significant activation in functional magnetic resonance imaging (fMRI): use of a cluster-size threshold. Magnetic Resonance in Medicine 33(5): 636-647.
Janca A, Bucholz K, Janca I (1997). Family history assessment module Version 5 Washington University School of Medicine: St Louis.
Jenkinson M, Bannister P, Brandy M, Smith S (2002). Improved optimization for the robust and accurate linear registration and motion correction of brain images. Neuroimage 17: 825-841.
Woolrich M, Ripley B, Brady M, Smith S (2001). Temporal autocorrelation in univariate linear modeling of FMRI data. Neuroimage 14:1370-1386.