Submission to Psychopharmacology
Oxytocin modulates hemodynamic responses to monetary incentives in humans
Brian J. Mickey 1,2,3, Joseph Heffernan 2, Curtis Heisel 2, Marta Peciña 1,2, David T. Hsu 1,2,4, Jon-Kar Zubieta 1,2,3, Tiffany M. Love 1,2,3
1Department of Psychiatry, University of Michigan Medical School, Ann Arbor, USA
2Molecular and Behavioral Neuroscience Institute, University of Michigan, Ann Arbor, USA
3Department of Psychiatry, University of Utah School of Medicine, Salt Lake City, USA
4Department of Psychiatry, Stony Brook School of Medicine, Stony Brook, New York, USA
Correspondence: Tiffany Love, PhD
Phone: 801-587-0160
Fax: 801-581-7109
3 tables, 6 figures, 1 supplement
Conflict of interest: During the 5-year period prior to submission, BJM served as a consultant to Alkermes, Inc., for work unrelated to this manuscript.
ClinicalTrials.gov registration NCT01722071 (
Supplementary Materials
Methods
Questionnaires.
Personality traits were characterized with the Revised NEO Personality Inventory [NEO PI-R (Costa and McCrae 1992)]. Emotional states were measured with the Positive and Negative Affect Schedule [PANAS (Watson and Clark 1994)], the Spielberger State-Trait Anxiety Inventory [STAI (Spielberger et al. 1983)], and the Brief Fear of Negative Evaluation (BFNE) scale.
NEO PI-R. The NEO PI-R is a 240-item self-report measure which provides a broad assessment of an individual’s personality. Participants are asked to indicate, on a scale of 1 to 5, the extent to which each item reflects their behavior and/or attitudes. The NEO PI-R reliably measures five major personality domains: Neuroticism, Extraversion, Openness, Agreeableness, and Conscientiousness.
PANAS. The PANAS is a 60-item self-report measure which can be used to assess current affective state. The PANAS offers two generally independent measures of affect: positive affect, defined as “the extent to which a person feels enthusiastic, active, and alert,” and negative affect, characterized as “the extent to which person feels displeased, distressed and aversive” (Watson et al. 1988). Items consist of series of mood descriptors and participants are asked to indicate, on a scale of 1 to 5, the extent to which each phrase reflects the way they currently feel.
STAI-State. The STAI-State is a brief 20-item self-report instrument which provides a measure of subjective state anxiety. Participants are given a series of statements and asked to indicate how each statement reflects the way they feel right now using a four-point scale.
BFNE. The BFNE scale is an instrument designed to measure fear of being negatively evaluated by others (Leary 1983). This scale consists of 12 statements which participants are asked to indicate, on a scale of 1 to 5, the extent to which each statement is characteristic of them. Higher scores are indicative of higher levels of fear of negative evaluation.
Results
Cardiac and respiratory rate analysis.
The BOLD signal in the brainstem may be vulnerable to pulsatile cardiac and respiratory artifacts, raising the possibility that the observed effects of oxytocin on midbrain BOLD responses could be due to indirect effects of oxytocin on heart or respiratory rate (see Gutkowska et al. 2014; Jankowski et al. 2012; Kemp et al. 2012, for discussion on oxytocin effects on cardiovascular activity). To evaluate this possibility, we compared mean heart and respiratory rates under oxytocin and placebo conditions. No significant effect of oxytocin was found for the 16 subjects with usable data (Table 1) indicating that this type of artifact is unlikely to explain our findings.
References:
Costa PT, McCrae RR (1992) Normal personality assessment in clinical practice: The NEO Personality Inventory. Psychological Assessment 4: 5-13.
Gutkowska J, Jankowski M, Antunes-Rodrigues J (2014) The role of oxytocin in cardiovascular regulation. Braz J Med Biol Res 47: 206-14.
Jankowski M, Gonzalez-Reyes A, Noiseux N, Gutkowska J (2012) Oxytocin in the heart regeneration. Recent Pat Cardiovasc Drug Discov 7: 81-7.
Kemp AH, Quintana DS, Kuhnert RL, Griffiths K, Hickie IB, Guastella AJ (2012) Oxytocin increases heart rate variability in humans at rest: implications for social approach-related motivation and capacity for social engagement. PLoS One 7: e44014.
Leary MR (1983) A Brief Version of the Fear of Negative Evaluation Scale. Pers Soc Psychol Bull 9: 371-375.
Spielberger CD, Gorssuch RL, Lushene PR, Vagg PR, Jacobs GA (1983) Manual for the State-Trait Anxiety Inventory. Consulting Psychologists Press, Inc.
Watson D, Clark LA (1994) PANAS-X Manual for the Positive and Negative Affect Schedule - Expanded Form, University of Iowa
Watson D, Clark LA, Tellegen A (1988) Development and validation of brief measures of positive and negative affect: the PANAS scales. J Pers Soc Psychol 54: 1063-70.
Supplementary Figure Legend
Fig S1. Task condition-specific effects of oxytocin on the hemodynamic response in the midbrain. The blood oxygenation level dependent (BOLD) signal in the midbrain (ventral tegmental area and substantia nigra) is shown versus time after presentation of each cue (beginning of the trial). Thick lines represent the mean across 18 subjects and thin lines indicate ± 1 standard error. Black dotted lines represent the placebo condition and red solid lines indicate the oxytocin condition. Each of the five task conditions (low-salience reward, high-salience reward, low-salience loss, high-salience loss, and neutral) are shown.
1
Tables
Table S1. Analysis of variance for post hoc linear mixed models of midbrain hemodynamic response time-course after treatment with oxytocin or placebo (6–12 second lag)
2 / df / p *Model B: Four conditions (neutral condition excluded)
Time / 44.8 / 1 / 2.2 x 10-11
Treatment / 0.44 / 1 / 0.51
Condition / 96.4 / 3 / <2.2 x 10-16
Time x Treatment / 10.8 / 1 / 9.9 x 10-4
Treatment x Condition / 5.97 / 3 / 0.11
Time x Condition / 28.5 / 3 / 2.9 x 10-6
Time x Treatment x Condition / 2.23 / 3 / 0.53
Model C: Salience and valence factors (neutral condition excluded)
Time / 44.8 / 1 / 2.2 x 10-11
Treatment / 0.44 / 1 / 0.51
Salience / 96.0 / 1 / <2.2 x 10-16
Valence / 0.042 / 1 / 0.84
Time x Treatment / 10.8 / 1 / 9.9 x 10-4
Time x Salience / 17.2 / 1 / 3.3 x 10-5
Time x Valence / 9.07 / 1 / 0.0026
Treatment x Salience / 1.31 / 1 / 0.25
Treatment x Valence / 2.86 / 1 / 0.091
Salience x Valence / 0.364 / 1 / 0.55
Time x Treatment x Salience / 1.37 / 1 / 0.24
Time x Treatment x Valence / 0.854 / 1 / 0.36
Time x Salience x Valence / 2.14 / 1 / 0.14
Treatment x Salience x Valence / 1.79 / 1 / 0.18
Time x Treatment x Salience x Valence / 0.0011 / 1 / 0.97
Model D: Low-salience reward only
Time / 5.12 / 1 / 0.024
Treatment / 7.18 / 1 / 0.0074
Time x Treatment / 2.79 / 1 / 0.095
Model E: High-salience reward only
Time / 62.7 / 1 / 2.5 x 10-15
Treatment / 0.0047 / 1 / 0.95
Time x Treatment / 7.85 / 1 / 0.0051
Model F: Low-salience loss only
Time / 0.309 / 1 / 0.58
Treatment / 0.474 / 1 / 0.49
Time x Treatment / 0.479 / 1 / 0.49
Model G: High-salience loss only
Time / 9.69 / 1 / 0.0018
Treatment / 0.169 / 1 / 0.68
Time x Treatment / 2.94 / 1 / 0.086
Model H: Neutral only
Time / 0.011 / 1 / 0.92
Treatment / 13.0 / 1 / 0.00031
Time x Treatment / 1.05 / 1 / 0.30
* Type II Wald chi-squared tests
Model B: Task condition is modeled as a factor with 4 levels: low-salience reward, high-salience reward, low-salience loss, and high-salience loss. Neutral trials are excluded;
Model C: Task condition is modeled as a cross of 2 factors, salience and valence. Salience has 2 levels, low and high. Valence has 2 levels, reward and loss. Response = time + treatment + salience + valence + time×treatment + time×salience + time×valence + treatment×salience + treatment×valence + salience×valence + time×treatment×salience + time×treatment×valence + time×salience×valence + treatment×salience×valence + time×treatment×salience×valence;
Model D: Only low-salience reward trials are included. Response = time + treatment + time×treatment;
Model E: Only high-salience reward trials are included. Response = time + treatment + time×treatment;
Model F: Only low-salience loss trials are included. Response = time + treatment + time×treatment;
Model G: Only high-salience loss trials are included. Response = time + treatment + time×treatment;
Model H: Only neutral trials are included. Response = time + treatment + time×treatment
1