Dear Editor
As part of the Centers for Intervention Development and Applied Research (CIDAR) initiative funded by the National Institute of Mental Health, galantamine and oxytocin were included in a three-arm, six week randomly controlled trial (ClinicalTrials.gov NCT01012167). Both medications are associated with mild side effect profiles (Hansen et al., 2008; Oya et al., 2016), are relatively inexpensive, and have shown efficacy in at least some studies (Lee et al., 2013). Unfortunately, data from the CIDAR trial failed to find significant improvements in negative symptom severity or cognitive functioning with either galantamine or oxytocin compared to placebo using standard clinical rating scales or neuropsychological test batteries (Buchanan et al., 2016). The present study reanalyzed these data using computerized measures of natural speech and facial expressions as the dependent variables; measures with potentially superior reliability, sensitivity and precision than clinical rating scales for measuring blunted affect and alogia (Cohen and Elvevag, 2014; Cohen et al., 2016; Cohen et al., 2013).
Participants in the study met criteria for schizophrenia or schizoaffective disorder as confirmed by structured clinical interview and a) were between 18 and 64 years old, b) were rated at moderate negative symptom severity or greater using a modified Scale for the Assessment of Negative Symptoms (Buchanan et al., 2007), and c) showed minimal levels of potential “secondary negative symptoms and extra-pyramidal symptoms (see Buchanan et al., 2016 for details). Fifty participants were included in the original study. Due to technical issues with audio or video recordings, data were available for 40 participants (15 Placebo; 10 oxytocin; 15 galantamine).
Vocal and facial expression was analyzed from a modified version of the Maryland Assessment of Social Competence (Bellack et al., 1994) using two conversational role-play situations enacted with female confederates whose responses were standardized and trained to be affiliative in their affect, rather than neutral. Data were averaged across the two situations. Assessments were administered at baseline and then repeated six weeks later. The Computerized assessment of Affect from Natural Speech protocol (Cohen et al., 2016) and FaceReader version 4.0, a commercially-available program developed by Noldus Information Technology (2010), were used to measure vocal and facial expressions respectively. Summary variables, selected based on recent studies (Cohen et al., 2016; Cohen et al., 2013), included: mean pause time (i.e., average voiceless epoch bounded by speech > 150ms in length), intonation (i.e., standard deviation of the fundamental frequency values computed within a voiced epoch [i.e., an “utterance”], then averaged across utterances), and emphasis (i.e., standard deviation of the volume computed within an utterance, then averaged across utterances), as were neutral, happy and negative (sum of sad, anger, scared) facial expressions. Extreme values were “winsorized” (i.e., replaced with values 3.5 SD from the overall means).
There were no statistically significant differences across treatment arms in sex, ethnicity, age, education, parental education or baseline negative symptom scores (p’s > .10). Treatment differences in average recording time and percentage of video frames analyzed by FaceReader also were also not statistically significant. In our main analyses, no statistically significant condition, treatment or interaction effects were noted for any of the six computerized measures. Interestingly, participants in the oxytocin arm showed a non-statistically significant increase in negative facial expressions pre- to post-treatment at a medium effect size level, compared to a negligible increase in the placebo and galantamine arms.
Consistent with data from symptom rating scales reported elsewhere from this medication trial (Buchanan et al., 2016), no significant treatment effects were observed. The reasons for this are unclear, though possible explanations include noncompliance or incomplete dosing (but see Lee et al 2013), insufficient power due to small sample size, short half-life and potential tolerance. Individuals in the oxytocin arm did show an increase in negative facial expressions across the two role-play tasks at a medium effect size (in contrast to negligible changes in the other arms). This finding is not entirely surprising given that the effects of oxytocin on prosocial behavior are context-dependent and can contribute to “antisocial” effects when an individual perceives themself or their “in group” members to be threatened in some manner (Kemp and Guastella, 2011). Given the confederates were established as unfamiliar people (e.g., neighbors), and in fact were largely unknown to the participants, they may have been viewed as “out group” members. Hence, oxytocin may have engendered feelings of suspicion or defensiveness in some fashion. However, confederates were trained and monitored to be affiliative during the role-plays. For this reason, and because the effects were not statistically significant, they should be viewed cautiously. In sum, the present study used highly sophisticated and sensitive computerized assessments, but failed to find significant improvements in vocal or facial expressions as a result of either galantamine or oxytocin.
Table 1.
Descriptive statistics between baseline and post-intervention treatment arms for computerized vocal and facial measures.
| Pre-Treatment | Post-Treatment | Condition F | Treatment F | Interaction F | Cohen’s da | |
|---|---|---|---|---|---|---|
| Vocal: Pause times | ||||||
| Placebo | 2694.19 (1486.06) | 2611.03 (1554.43) | 0.67 | 0.76 | 0.67 | −0.05 |
| Oxytocin | 2303.19 (1306.69) | 1958.76 (1150.16) | −0.28 | |||
| Galantamine | 2434.51 (1049.42) | 2300.88 (951.35) | −0.13 | |||
| Vocal: Intonation | ||||||
| Placebo | 2.22 (0.87) | 1.87 (0.48) | 2.15 | 2.04 | 2.15 | −0.52 |
| Oxytocin | 2.53 (1.30) | 1.93 (0.63) | −0.62 | |||
| Galantamine | 2.51 (1.17) | 2.67 (0.84) | 0.16 | |||
| Vocal: Emphasis | ||||||
| Placebo | 4.32 (3.93) | 3.07 (1.23) | 2.05 | 0.26 | 2.05 | −0.48 |
| Oxytocin | 4.84 (4.34) | 3.54 (0.96) | −0.49 | |||
| Galantamine | 4.02 (1.70) | 4.01 (1.31) | −0.01 | |||
| Facial: Neutral | ||||||
| Placebo | 0.32 (0.11) | 0.30 (0.15) | 0.06 | 0.06 | 0.06 | −0.15 |
| Oxytocin | 0.32 (0.15) | 0.33 (0.19) | 0.06 | |||
| Galantamine | 0.32 (0.13) | 0.31 (0.12) | −0.08 | |||
| Facial: Happy | ||||||
| Placebo | 0.08 (0.09) | 0.12 (0.11) | 0.14 | 0.68 | 0.40 | |
| Oxytocin | 0.06 (0.08) | 0.07 (0.10) | 0.14 | 0.11 | ||
| Galantamine | 0.10 (0.11) | 0.08 (0.09) | −0.20 | |||
| Facial: Negative | ||||||
| Placebo | 0.16 (0.13) | 0.16 (0.13) | 1.10 | 0.08 | 1.10 | 0.00 |
| Oxytocin | 0.14 (0.11) | 0.20 (0.10) | 0.57 | |||
| Galantamine | 0.16 (0.13) | 0.18 (0.12) | 0.16 | |||
d values computed pre to post within each treatment arm.
Acknowledgments
Funding
The study was supported by NIMH grant 1P50 MH082999: MPRC Centers for Intervention Development and Applied Research (CIDAR) (P.I.: WTC). These findings were presented at the International Congress of Schizophrenia Research in Colorado Springs, CO in April of 2015 and the European College of Neuropsychopharmacology in August of 2015.
The authors thank the research and treatment teams at the Maryland Psychiatric Research Center and Spring Grove Hospital Center for their help in all study aspects. They thank Jennifer Osing and Jared Linthicum for coordinating all study procedures; Drs. Heidi J Wehring, Douglas Boggs and Bernard Fischer for assistance with the FDA IND; and Frank Blatt, Pharm.D. for the preparation of galantamine and dispensing and monitoring of all study medications.
Footnotes
Contributors. Alex S Cohen was the primary author for this project, conducted the statistical analyses and wrote the bulk of the manuscript. All other authors helped manage the literature searches, interpret the findings and provided conceptual material to the planning and presentation of this project. All authors contributed to and have approved the final manuscript.
Conflict of Interest and Financial Support Statement: Robert W. Buchanan: DSMB member: Consultant: AbbVie; Advisory Board: AbbVie; Amgen; Boehringer Ingelheim-RCV; EnVivo; Lundbeck; Takeda; Deanna L. Kelly: Advisory Board: Otsuka, XOMA and Janssen; Elaine Weiner: no competing interests or financial support to disclose; James M. Gold: Consultant: Amgen and Hoffman LaRoche; and receives royalty payments from the BACS.; Gregory P. Strauss: receives royalties and consultation fees from ProPhase LLC in connection with the commercial use of the Brief Negative Symptom Scale and other professional activities; Maju M. Koola: no competing interests or financial support to disclose; Robert P. McMahon: no competing interests or financial support to disclose; and William T. Carpenter: Consultant: HealthAnalytics and Pharmagenesis; Advisory Board: Allergen and Teva.
Publisher's Disclaimer: 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 citable 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.
Contributor Information
Alex S. Cohen, Louisiana State University, Department of Psychology
Kyle R. Mitchell, Louisiana State University, Department of Psychology
Gregory P. Strauss, Binghamton University, Department of Psychology
Jack J. Blanchard, University of Maryland, College Park, Department of Psychology
Robert W. Buchanan, University of Maryland School of Medicine, Maryland Psychiatric Research Center
Deanna L. Kelly, University of Maryland School of Medicine, Maryland Psychiatric Research Center
James Gold, University of Maryland School of Medicine, Maryland Psychiatric Research Center.
Robert P. McMahon, University of Maryland School of Medicine, Maryland Psychiatric Research Center
Heather A. Adams, University of Maryland School of Medicine, Maryland Psychiatric Research Center
William T. Carpenter, University of Maryland School of Medicine, Maryland Psychiatric Research Center
References
- 1.Buchanan RW, Javitt DC, Marder SR, Schooler NR, Gold JM, McMahon RP, Uriel Heresco-Levy M, Carpenter WT. The Cognitive and Negative Symptoms in Schizophrenia Trial (CONSIST): the efficacy of glutamatergic agents for negative symptoms and cognitive impairments. Am J Psychiatry. 2007;164(10):1593–1602. doi: 10.1176/appi.ajp.2007.06081358. [DOI] [PubMed] [Google Scholar]
- 2.Buchanan RW, Kelly DL, Strauss EWJMGGP, Koola MM, McMahon RP, Carpenter WT. A Randomized Clinical Trial of Oxytocin and Galantamine for the Treatment of Negative Symptoms and Cognitive Impairments in People with Schizophrenia. 2016 doi: 10.1097/JCP.0000000000000720. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Cohen AS, Elvevag B. Automated computerized analysis of speech in psychiatric disorders. Curr Opin Psychiatry. 2014;27(3):203–209. doi: 10.1097/YCO.0000000000000056. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Cohen AS, Mitchell KR, Docherty NM, Horan WP. Vocal expression in schizophrenia: Less than meets the ear. J Abnorm Psychol. 2016;125(2):299–309. doi: 10.1037/abn0000136. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Cohen AS, Morrison SC, Callaway DA. Computerized facial analysis for understanding constricted/blunted affect: initial feasibility, reliability, and validity data. Schizophr Res. 2013;148(1–3):111–116. doi: 10.1016/j.schres.2013.05.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Hansen RA, Gartlehner G, Webb AP, Morgan LC, Moore CG, Jonas DE. Efficacy and safety of donepezil, galantamine, and rivastigmine for the treatment of Alzheimer’s disease: a systematic review and meta-analysis. Clin interv aging. 2008;3(2):211. [PMC free article] [PubMed] [Google Scholar]
- 7.Kemp AH, Guastella AJ. The role of oxytocin in human affect a novel hypothesis. Curr Dir Psychol Sci. 2011;20(4):222–231. [Google Scholar]
- 8.Lee MR, Wehring HJ, McMahon RP, Linthicum J, Cascella N, Liu F, Bellack A, Buchanan RW, Strauss GP, Contoreggi C, Kelly DL. Effects of adjunctive intranasal oxytocin on olfactory identification and clinical symptoms in schizophrenia: results from a randomized double blind placebo controlled pilot study. Schizophr Res. 2013;145(1–3):110–115. doi: 10.1016/j.schres.2013.01.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Oya K, Matsuda Y, Matsunaga S, Kishi T, Iwata N. Efficacy and safety of oxytocin augmentation therapy for schizophrenia: an updated systematic review and meta-analysis of randomized, placebo-controlled trials. European archives of psychiatry and clinical neuroscience. Eur Arch Psychiatry Clin Neurosci. 2016;266(5):439–50. doi: 10.1007/s00406-015-0634-9. [DOI] [PubMed] [Google Scholar]
- 10.Noldus Information Technology. FaceReader 4.0. 2010. [Google Scholar]