Intranasal Oxytocin and Maternal Neglect

The investigators plan to conduct fMRI scanning with a group of mothers who are blindly and randomly assigned either intranasal oxytocin or a placebo.

Aim 1: To determine whether intranasal oxytocin (IN-OT) affects parental brain responses to infant cues, as measured during functional MRI (fMRI) scanning.

Hypothesis 1 (i): Compared to placebo, mothers who receive intranasal oxytocin will show greater activation of mesocorticolimbic dopamine reward regions in the brain, including the ventral striatum and the medial prefrontal cortex, when viewing their own vs. unknown infant faces during fMRI scanning. Hypothesis 1 (ii): Compared to placebo, fathers will show greater activation of social-cognitive circuits, including the dorsal prefrontal cortex (dPFC).

Aim 2: To determine whether IN-OT affects parental behavioral responses to infant cues, as measured by videotaped parent-infant interaction procedures.

Hypothesis 2 (i): Compared to placebo, parents who receive intranasal oxytocin will score higher on an overall measure of maternal sensitivity, during a free play parent-infant interaction procedure (CARE-Index). Hypothesis 2 (ii): Compared to placebo, mothers who receive intranasal oxytocin will demonstrate more frequent emotionally contingent responses with their infant during the recovery phase of a modified Still-Face procedure.

Aim 3: To determine whether attachment security interacts with the effect of IN-OT on maternal brain and behavioral responses to infant cues.

Hypothesis 3 (i): Compared to secure (Type B) mothers, mothers with insecure/dismissing (Type A) attachment will have a greater brain and behavioral response to IN-OT (as described in Hypotheses 1 and 2), with a 2-way interaction effect seen between treatment and attachment groups. Hypothesis 3 (ii): Compared to Type B mothers, mothers with insecure/preoccupied (Type C) attachment will have a reduced brain and behavioral response to IN-OT (as described in Hypotheses 1 and 2), with a 2-way interaction effect seen between treatment and attachment groups.

Aim 4: To determine whether brain reward activation is associated with other indirect measures of emotional neglect in mothers.

Hypothesis 4 (i): Maternal brain reward system activation, both with and without IN-OT, will be positively correlated with 1) maternal sensitivity (as measured by the CARE-Index), 2) the contingency of maternal responses to infant behavioral cues.

Initial enrollment: Recruitment will occur during the postpartum period. In addition to recruiting from the medical record, the investigators plan to execute other recruiting measures in the form of advertising. Parent(s) will be recruited via newspaper, internet, public postings, and mass mailings. Once eligibility is confirmed, subjects will be scheduled for a study visit where they will be informed of confidentiality and consented before participating in any study activities.

Visit 1: 4 Months Post-Partum - Adult Attachment Interview (AAI). During this visit, each enrolled woman will participate in a modified version of the Adult Attachment Interview (AAI), a semi-structured 1½-2 hour-long interview involving specified questions and follow-up inquiries relating to childhood relationships with attachment figures. The investigators will also collect sociodemographic, breastfeeding data, and screening information for depression. The Edinburgh Postnatal Depression Scale (EPDS) will be repeated during subsequent visits in order to calculate mean scores.

Visits 2A\&B and 3A\&B: Intranasal Oxytocin vs. Placebo Administration In this double-blind cross-over placebo-controlled trial of intranasal oxytocin, maternal brain and behavioral responses to infant cues will be assessed within \& between subjects. Each participant will be administered both the active (oxytocin) and inactive (placebo) nasal spray just prior to separate behavioral or fMRI scanning sessions, with order of administration balanced across subjects and between attachment groups. All involved will be blinded to the identity of the oxytocin or placebo sprays. Before each visit, parent(s) will be asked to abstain from tobacco, food and drink (except water) for at least 2 hours prior to the visit. The investigators will attempt to schedule visits during the latter half of the mother's menstrual cycle, to minimize potential confounding from fluctuations in estrogen levels. Mothers without regular menstrual cycles will be scheduled one month apart. Mothers who are still breastfeeding will be asked to feed their infant at least one hour prior to the visit. For the four visits involving intranasal drug administration (two behavioral testing visits and two fMRI scanning visits), the following protocol will apply: participants will be given a urinary pregnancy test prior to the start of each visit. If a mother tests positive, she will be excluded from the study due to theoretical concern about possible effects of oxytocin on uterine contraction. Due to the fact that the pregnancy test may still not detect an early pregnancy, the mothers will also be asked to abstain from unprotected sexual intercourse during the 2 weeks prior to the visit, and will confirm, in signing the consent form, that there is no substantial chance of a current pregnancy. Just prior to drug administration, the parent(s) will complete the Positive and Negative Affect Scale, to rate her current emotions, for comparison to post-spray. The parent(s) will then self-administer a dose of either oxytocin (3 puffs per nostril [4 IU per puff] = 24 IU total) or a placebo spray that contains only the inactive ingredients of the oxytocin solution. Both experimenters and subjects will be blind to the treatment they are receiving. A stopwatch will be started at the moment the subjects begins intranasal administration, so that the behavior assessment or fMRI scanning can begin exactly 50 minutes later. Most other studies of intranasal oxytocin have used a 50 minutes delay time, based on cerebrospinal fluid studies of other intranasally administered neuropeptides, such as the analogous hormone vasopressin. Forty minutes after drug administration, the PANAS will be repeated.

Visits 2A\&B: 5-6 Months Postpartum - Behavioral Response

1. On arrival, the parent(s) and infant will be escorted into a behavioral observation room at the Clinical Nutrition Research Center, where the infant will be introduced to some developmentally appropriate toys. Questionnaires completed at home will be reviewed (Infant Behavior Questionnaire [IBQ], Parenting Stress Index [PSI] and demographics and breastfeeding questionnaire), and if incomplete will be completed at this time.
2. Each parent will then self-administer the nasal spray, which has been previously randomized by the investigational pharmacy to be either oxytocin or placebo. During the following waiting period, the Bayley Scales of Infant Development screener will be completed with the infant, to exclude developmental delays in areas of fine motor, gross motor, language or social development. This will be performed by a research associate who has been trained in the use of this standardized tool.
3. Fifty minutes after the nasal spray is administered, each parent will rejoin their infant to each participate in a videotaped "free play" interaction on the floor for 3 minutes (described below). The recording will later be coded for dyadic interactions using a 14-point sensitivity scale in the CARE-Index.
4. A modified still-face procedure will then be conducted during the next 3 minutes(as described below).
5. The parent(s) will then be taken to another area of the building, while the infant is videotaped to obtain face images that will be used during the subsequent scanning visits. During the brief separation period, the infant will be secured in a mounted car seat, and videotaped while using age appropriate toys to elicit smiling (and neutral) facial expressions. This should last no more than 10 minutes. Crying faces will also be elicited, if necessary, by briefly leaving the infant alone in the room, secured in the car seat, while the video recorder is running. The infant will be observed at all times from behind a one-way mirror, and only left to cry for around 35-40 seconds. Infant cry will also be recorded at this time.
6. Infant face images with various degrees of affect-happy, neutral and sad-will then be extracted from the digital video recording, for use in the subsequent fMRI procedure (Visit 3). The images will be standardized for size, orientation and background using Adobe Photoshop. All of the face images will be rated on degree of affect using a 5-point adaptation of the Self-Assessment Manikin (SAM), categorizing each image into one of five affect groups: very happy, happy, neutral, sad or very sad. These faces will be matched on age, sex, ethnicity and degree of affect with a large database of "unknown" infant faces.
7. Approximately one month later, at a comparable time in the menstrual cycle, the basic procedure will be repeated for each subject, except that those who received intranasal oxytocin will be given placebo, and vice versa.

Visit 3A\&B: 8-9 Months Postpartum - Brain Response

1. Three months after the first behavioral response session, each parent will attend two scanning sessions at the Human Neuroimaging Laboratory, Baylor College of Medicine. Intranasal oxytocin or placebo will again be administered as described above. Infant face images from Visit 3A will be used in Visit 4A scanning session, and Visit 3B with 4B, to ensure that the repeat scanning session uses comparable but unique face images. The images from both sessions will be matched on degree of facial affect. Prior to intranasal drug administration on the first scanning visit, the parent(s) will complete the Positive and Negative Affect States (PANAS) questionnaire.
2. During the waiting period, the parent(s) will complete questionnaires.
3. Fifty minutes after the administration of the nasal spray, the parent(s) will participate in the functional MRI scanning session, passively viewing a series of unique infant face images and hearing cries of her own infant and of an unknown infant. Using an event-related fMRI design, randomly presented images will be viewed for 2 seconds, with a random inter-stimulus interval of 2, 4 or 6 seconds (Fig 3). The 60 images will be equally divided into 3 affect groups - happy, neutral or sad, with the intensity of happy and sad affect balanced between the "own" and "unknown" faces. The order of the images from each of the 6 groups (Own-Happy [OH], Own-Neutral [ON], Own-Sad [OS], Unknown-Happy [UH], Unknown-Neutral [UN], Unknown-Sad [US]) will be pseudo-randomized within the run, but not between subjects.
4. All imaging will be performed using a 3 Tesla Siemens MRI scanner. Visual images will be generated using a projector, and presented to the parent(s) via an overhead mirror display. Regional brain activation will be assessed by measuring changes in blood-oxygen-level-dependent functional MRI signal (BOLD-fMRI).
5. Subjects will participate in a single whole-brain functional run of around 185 scans each.
6. High-resolution T1-weighted structural images will then be acquired.
7. After the scanning session, each parent will be asked to rate each of the infant face images on how they thought the infant was feeling, as well as their own feelings of pleasure or arousal, using an adaptation of the Self-Assessment Manikin.
8. After the parent(s) has completed the questionnaires, 2 out of 4 randomly selected modules of the Bayley Scales of Infant Development screener will be completed with the infant.
9. The parent(s) will then rejoin their infant to participate in a videotaped "free play" interaction on the floor for 3 minutes. The recording will later be coded for dyadic interactions using a 14-point sensitivity scale in the CARE-Index.