Explicit and implicit caregiving interests in expectant fathers: Do endogenous and exogenous oxytocin and vasopressin matter?

Infant Behavior & Development 41 (2015) 26–37

Contents lists available at ScienceDirect

Infant Behavior and Development

Explicit and implicit caregiving interests in expectant fathers: Do endogenous and exogenous oxytocin and vasopressin matter? Celina C.C. Cohen-Bendahan a,∗ , Roseriet Beijers a , Lorenz J.P. van Doornen b , Carolina de Weerth a a b

Department of Developmental Psychology – Behavioural Science Institute, Radboud University Nijmegen, The Netherlands Department of Clinical and Health Psychology, Utrecht University, The Netherlands

a r t i c l e

i n f o

Article history: Received 7 December 2014 Received in revised form 6 March 2015 Accepted 10 June 2015 Keywords: Fathers Caregiving Pregnancy Virtual reality Oxytocin Vasopressin Hormones

a b s t r a c t Caregiving interest in men (N = 46) during the third trimester of their partner’s pregnancy was examined. The study included both explicit and implicit measures of caregiving interest, assessments of basal urinary concentrations of oxytocin and vasopressin, and exogenous (intranasal) application of these hormones. Compared to control men (N = 20), fathers-to-be reported more interest in direct care for children. In an immersive virtual environment, fathers-to-be, in comparison to control men, stood closer to and tended to spend more time looking at the baby-related avatars, and stood further away and tended to spend less time looking at non-baby-related avatars. Basal oxytocin and vasopressin were not related to caregiving interest in fathers-to-be, and were not different from control men. When vasopressin was administered, fathers-to-be invested more time watching the babyrelated avatars compared to control men. No effects were found of exogenous oxytocin on the behavior of fathers-to-be and control men in the immersive virtual environment. These results point in the direction of an adjustment of fathers-to-be for fatherhood, both consciously and unconsciously, and support the possible role of vasopressin in human behavior in the transition to fatherhood. © 2015 Elsevier Inc. All rights reserved.

1. Introduction In many cultures around the world extensive paternal care is fairly common (Marlowe, 2000). In contrast, in Western societies fathers have historically dominated the workforce and been expected to have little involvement in child caregiving responsibilities. Over the last decades, an increasing number of mothers have continued working after childbirth and this has brought changes in attitudes toward the fatherhood role (Cabrera, Tamis-LeMonda, Bradley, Hofferth, & Lamb, 2000). Apparently, there is a transitional shift toward men becoming recognizable caregivers (Kentner, Abizaid, & Bielajew, 2010; Lamb, 2010). Moreover, next to the maternal contribution, paternal involvement is recognized to be also important in childhood development. Active and regular engagement of the father with the child predicts a range of positive child outcomes, including less behavioral problems, less psychological problems, and enhanced cognitive development (Sarkadi,

∗ Corresponding author at: Department of Developmental Psychology – Behavioural Science Institute, Radboud University Nijmegen, PO Box 9104, 6500 HE Nijmegen, The Netherlands. Tel.: +31 24 3612637. E-mail address: celina [email protected] (C.C.C. Cohen-Bendahan). http://dx.doi.org/10.1016/j.infbeh.2015.06.007 0163-6383/© 2015 Elsevier Inc. All rights reserved.

C.C.C. Cohen-Bendahan et al. / Infant Behavior & Development 41 (2015) 26–37

27

Kristiansson, Oberklaid, & Bremberg, 2008). Nevertheless, there is significant variation among men in their levels of paternal care. While some fathers show more interest and spend more time caregiving for their children, other fathers do not (Hewlett, 1991). As most investigations focused on maternal care, and only a few studies investigated the hormonal correlates underlying paternal care (Gettler, McDade, Feranil, & Kuzawa, 2011; Gray, Parkin, & Samms-Vaughan, 2007), this will be studied in the present study. Caregiving interest will be examined in men who are about to become a father. Do males during their partners’ pregnancy show more interest in infant caregiving as compared to childless male controls? In addition, do these expectant fathers show basal hormonal differences that are related to their caregiving interests? Moreover, we aim to investigate whether caregiving interest can be enhanced in expectant fathers and controls when exogenous hormones are administered. Hormones have an important role in affiliative behavior, pair-bonding and also caregiving behavior (Heinrichs & Domes, 2008; MacDonald & MacDonald, 2010; Rilling, 2013). In animals, the changes in hormones involved in birth and lactation, including oxytocin, are also implicated in maternal behavior (Bosch & Neumann, 2012; Fernandez-Duque, Valeggia, & Mendoza, 2009). For example, increasing concentrations of oxytocin during pregnancy have been found to be related to higher maternal-fetal bonding (Levine, Zagoory-Sharon, Feldman, & Weller, 2007) and higher quality of human mother–infant interactions (Feldman, Gordon, Schneiderman, Weisman, & Zagoory-Sharon, 2010; Feldman, Gordon, & Zagoory-Sharon, 2011; Gordon, Zagoory-Sharon, Leckman, & Feldman, 2010). The hormonal correlates of paternal care have not been studied as extensively as those of maternal care (Bridges, 2010; Gettler et al., 2011; Gray et al., 2007). Nevertheless, there are indications that changes in hormone patterns observed in expectant human fathers residing with their partner, may prime the onset of paternal behavior (Storey, Walsh, Quinton, & Wynne-Edwards, 2000). Moreover, in paternal animals a similar pattern has been shown, together with an absence of similar hormonal changes in non-paternal animal species (e.g., Reburn & Wynne-Edwards, 1999). Although males are not directly affected by pregnancy-related hormones, hormones released in response to mating and cues from pregnant partners might facilitate male parental care (Brown, 1993). Indeed, there is some evidence in humans that male hormone concentrations change during their partner’s pregnancy. For example, Storey et al. (2000) observed elevated prolactin and cortisol levels in expectant fathers immediately prior to birth, while Berg and Wynne-Edwards (2001) reported higher estradiol levels and reduced testosterone levels during the last few months of their partner’s pregnancy. It is less clear whether male oxytocin concentrations also change during pregnancy and are related to a possible increase in caregiving interest in men who are about to become fathers. In addition to oxytocin, paternal vasopressin concentrations might also change during pregnancy. While both oxytocin and vasopressin have been widely recognized as important for affiliation, and the architecture of these two hormones is very similar (i.e. 7 of 9 amino acid positions are identical; Bosch & Neumann, 2012; Carter, 1998), vasopressin is suggested to be even more affiliated with paternal behavior than oxytocin (Carter, 2007; Storm & Tecott, 2005; Taylor, Saphire-Bernstein, & Seeman, 2010; Wang, Liu, Young, & Insel, 2000; Wynne-Edward, 2001). For example, Taylor et al. (2010) showed that relationship distress was more associated with oxytocin in women and with vasopressin in men. They suggest that vasopressin may function in men as oxytocin functions in women. To date, most studies focused on how endogenous variations in hormone concentrations correlate with social behavior. Within this context, however, it should be noted that there are several confounding factors, including the release of other hormones, for which it is difficult to control (Heinrichs & Domes, 2008). Fortunately, neuropharmacological research has shown that neuropeptides gain access to the human brain after intranasal administration (Born, Pietrowsky, & Fehm, 1999; Neumann, Maloumby, Beiderbeck, Lukas, & Landgraf, 2013; Pietrowsky, Strüben, Mölle, Fehm, & Born, 1996) for studying the central nervous effects of oxytocin and vasopressin (Heinrichs & Gaab, 2007; MacDonald & MacDonald, 2010). Studies using oxytocin and vasopressin nasal sprays have already shown to promote social cognition and the interpretation of social communication, possibly representing an enhanced readiness to show social approach behavior and empathy (Heinrichs & Domes, 2008). In this line, it can be hypothesized that caregiving interest, as a manifestation of social behavior, can also be manipulated when oxytocin and vasopressin are administered. In order to be better able to unravel the underlying biological mechanisms of human paternal care, studies are needed involving intranasal administration of hormones in double-blind, placebo-controlled designs (MacDonald & MacDonald, 2010). Caregiving interest in expectant fathers may be influenced by hormones and cues from pregnant partners (Hirschenhauser, Frigerio, Grammer, & Magnusson, 2002). In turn, caregiving interest may be related to how fathers-tobe perceive environment relevant cues, especially those related to infants and the paternal role. In social psychological research, two types of perception processes are distinguished (Lansu, Cillessen, & Karremans, 2012). Explicit perceptions or associations are deliberate, controlled, and under awareness. Implicit perceptions or associations are non-deliberate, automatic, and often without awareness. These explicit and implicit perceptions do not need to be concordant, as explicit perceptions for example can be subject to socially desirable answering (Paulhus, 1991). Taking into account that expectant fathers are going through psychological as well as physiological changes whilst approaching parenthood, we hypothesize that expectant fathers will show more explicit as well as implicit caregiving interest. While explicit perceptions will be studied through self-report, implicit perceptions will be studied by observing the behavior of expectant fathers in a virtual reality setting containing baby cues. Immersive Virtual Environment (IVE) is a technique that permits participants to move around freely in an immersive, three-dimensional computer-generated environment. While the participant walks around in the test room, he views the virtual environment designed for the test room

28

C.C.C. Cohen-Bendahan et al. / Infant Behavior & Development 41 (2015) 26–37

through the head-mounted display. The IVE allows for (1) optimal control over the experimental conditions, and (2) unobtrusive and accurate measurement of dependent variables (e.g., approach behavior, as a measure of implicit perceptions) (Rinck, Kwakkenbos, Dotsch, Wigboldus, & Becker, 2010). In the present study, we expect participants to respond to avatars (a graphical representation of a character in a three-dimensional form) according to their interest in care for infants and children. It is hypothesized that men who are more interested in caregiving, will spend more time in the vicinity and looking at baby avatars than at a female avatar or at moving objects (toy). To summarize, we aim to examine the caregiving interest in men who are about to become a father for the first time, and compare their caregiving interest to men who are not about to become fathers. Caregiving interest will be measured with explicit (questionnaire) and implicit measures (IVE). We hypothesize that, during the third trimester of their partner’s pregnancy, men will report more interest in caregiving as compared to childless controls, and that this increased interest will be correlated with increased basal urinary oxytocin and vasopressin concentrations. Furthermore, we expect that expectant fathers will show more implicit caregiving interest, i.e. increased approach toward the baby-related avatars and less approach toward the non-baby-related avatars, compared to the controls. Moreover, we aim to investigate whether caregiving interest can be experimentally enhanced in expectant fathers and control males by administering exogenous oxytocin and vasopressin via nasal sprays. We expect men to spend more time and to be more in the vicinity of baby-related avatars when oxytocin and vasopressin are administered. 2. Methods 2.1. Participants Forty-six expectant nulliparous fathers and twenty control men participated in this study. All men were cohabitating with their partners and did not have children; the control men’s partners were not expecting a child, and they had the desire to become fathers at some future time, but not at the current moment. The participants were recruited in several ways; via midwife practices, door-to-door flyers, and posters in supermarkets, libraries and places on and nearby the university campus. Exclusion criteria were medical or psychiatric illness, use of medication, substance abuse, and heavy smoking or drinking. All participants were healthy volunteers who received an incentive of 50 euros for their participation. The study was carried out in accordance with the Declaration of Seoul principles and Ethical approval was provided by the Medical Ethics Committee (protocol number 09-244), and by Central Committee on Research Involving Human Subjects (CCMO) as the Competent Authority for the review of clinical trials in The Netherlands (protocol number NL28361.041.09). All subjects gave written, informed consent and were informed of their right to discontinue participation at any time. Demographic characteristics are provided in Table 1. The expectant fathers did not differ in educational level, nor in the amount of former experience with children (all p’s > .30). However, fathers-to-be were on average 7 years older than the controls (33.5 years and 26.4 years, respectively). Correlational analyses revealed that the age of the participant was not associated with any explicit or implicit caregiving interest or basal hormones (all p > .15), therefore we did not consider age as a confounder for our further analyses. 2.2. Procedure Before the lab visit, all participants completed an online questionnaire at home with questions on the following topics: demographics, personality traits, health, interests and attitudes toward children and parenthood. Thereafter, participants were invited to visit the lab weekly for three consecutive weeks. The first visit of the expectant fathers was planned at 32 week gestation (Mweek1 = 32.4, SDweek1 = 1.7; Mweek2 = 33.4, SDweek2 = 1.5; Mweek3 = 34.4, SDweek3 = 1.5). Participants were instructed to abstain from alcohol and caffeine on the evening before and on the day of testing, and from food and drinks

Table 1 Descriptive statistics of demographic characteristics. Mfathers (SD)

Demographic characteristics Age (years) Educational level (%) Secondary education College/University Experience with children (%) A lot Some Little No **

p < .01.

Mcontrols (SD)

33.49 (4.72)

26.35 (3.73)

22.3% 77.8%

45% 55%

6.7% 33.3% 55.6% 4.4%

0% 20% 75% 5%

Group differences t/X2

p-value

t(63) = 5.974 X2 (5,N=65) = 7.421

.000** .919

X2 (3,N=65) = 3.035

.386

C.C.C. Cohen-Bendahan et al. / Infant Behavior & Development 41 (2015) 26–37

29

(except water) for 2 h before the beginning of the lab visit. All visits took place in the late afternoon and evening, starting between 4:00 and 7:00 pm, in order to control for possible diurnal fluctuations in hormone levels. The participant was given a short introduction regarding the course of the lab visit and a questionnaire to be filled out. After approximately 25 min, the intranasal administration took place in a double-blind, intra-subject, counterbalanced crossover design. This means that each week, one of the three different intranasal sprays was administered to the participant in a random order. The order of the intranasal sprays was counter-balanced between the participants, and the experimenter and participants were blind to the intranasal spray given. Forty-five minutes after the intranasal spray was administrated the experimenter accompanied the participant to the Radboud Immersive Virtual Environment research lab (RiverLab) for the IVE experiment. They received a short instruction and were immersed in an IVE for two and a half minutes. We piloted the virtual environment and reached the conclusion that this is the right time for a person to have enough time to look around without getting bored. Participants were instructed to take a quick and equal glance of all the avatars by having them stand in the center of the room and make a circle of 360◦ , starting at a random virtual position. 2.3. Hormone administration Participants received intranasal oxytocin (OT; 24 IU/ml; Syntocinon, Novartis Pharma), synthetic vasopressin (AVP; 80 IU/ml; Desmopressine-acetaat Actavis), or placebo (saline PCH, Pharmachemie B.V.) in a double-blind, intra-subject, counterbalanced crossover design. The hormonal dosages and the timing of the experiment (45 min) were based on prior studies (see for example MacDonald & MacDonald, 2010). Effects of the oxytocin and vasopressin nasal sprays manifest within 10 min after administration and last for at least 80 min (e.g., Born et al., 2002). 2.4. Measures 2.4.1. Basal oxytocin and vasopressin concentrations Several days before each lab visit, participants received materials and instructions on how to collect urine samples at home. Every visit, the participants brought one urine sample taken on the morning of the lab visit (i.e. first voiding of the morning). These samples were immediately stored at −18 ◦ C. The urine samples were used to assess the basal levels of oxytocin and vasopressin. Levels of oxytocin were determined using the Oxytocin EIA kit (Enzo Life Sciences Inc., USA), a competitive immunoassay for the quantitative determination of oxytocin. Levels of vasopressin were determined using the Arg-Vasopressin EIA kit (Enzo Life Sciences Inc., USA), a competitive immunoassay for the quantitative determination of vasopressin. Both the vasopressin and oxytocin kits have been validated in various species and materials, including urine (e.g., OT: Moscovice & Ziegler, 2012; AVP: Gray et al., 2007). The kits were validated by our lab for the use of urine by testing urine of 20 healthy volunteers undiluted and in a 1:1 dilution with assay diluent. This gave a good correlation with r2 = 0.98 (rc = 0.98). Furthermore, 2 urine samples were 1:1 spiked with the standards samples and measured. This also resulted in a linear relationship with good correlation (r2 = 0.98). The values for the volunteers were all within the measuring range, and mean differences between the duplo measurements of both urine samples were 8.9 and 10.3%. The sensitivity of the methods was 3.39 and 11.7 pg/mL, respectively for vasopressin and oxytocin. The intra-assay CV was 10.2% at the level of 6.31 pg/mL for vasopressin, and 9.1% at the level of 21.4 pg/mL for oxytocin. The inter-assay CV was 8.5% at the level of 5.54 pg/mL for vasopressin, and 14.5% at the level of 19.3 pg/mL for oxytocin. Urinary creatinine concentrations were measured using the Jaffé method. The basal urinary oxytocin and vasopressin levels were expressed as the oxytocin to creatinine ratio and the vasopressin to creatinine ratio, respectively. The weekly basal urinary oxytocin levels did not correlate with each other (r’s < .12, p’s > .20). For weekly basal urinary vasopressin levels, weeks 1 and 3 were correlated r(51) = .38 (p < .01), and weeks 2 and 3 were correlated r(52) = .59 (p < .001); but weeks 1 and 2 were not correlated r(56) = .14 (p > .29). As the oxytocin levels were not correlated, we refrained from calculating mean basal urinary hormone levels over all weeks. Instead, the weekly basal hormone levels were used to examine possible differences between fathers-to-be and control participants on the morning of each lab visit. In addition, in order to test the association between basal hormone levels and the implicit and explicit caregiving interests in fathers-to-be, we used the hormone levels corresponding to the placebo condition of the lab visit. 2.4.2. Explicit caregiving interest To identify care interest for infants and young children, a measure of explicit caregiving interest was conducted, in which participants indicated on a five- or six-point Likert scale their interest in different caregiving activities (Cohen-Bendahan, van Doornen, & de Weerth, 2014). This measure consists of 25 items in three different domains of explicit care, namely direct care (e.g., ‘how often do you think you will be changing a diaper?’, Cronbach’s ˛ = .91), indirect care (e.g., ‘how often do you think you will be shopping for the child’s toys and clothes?’, Cronbach’s ˛ = .78), and play (e.g., ‘how often do you think you will be reading stories to your child?’, Cronbach’s ˛ = .79). Z-scores were calculated for each item. 2.4.3. Implicit caregiving interest The immersive virtual environment (IVE) was displayed using a stereoscopic head-mounted display (HMD); a helmet-like apparatus with internal computer displays that cover 60◦ diagonal of the participant’s field of view. During the walk in the virtual environment, the participant views the room through the HMD. A sensor on top of the HMD, signals the participant’s

30

C.C.C. Cohen-Bendahan et al. / Infant Behavior & Development 41 (2015) 26–37

position and orientation to a computer that continuously renders images of the virtual environment at 33 Hz, consistent with the participant’s physical position in the room. The participants could walk freely in an area of 4.5 m × 5.5 m, while the avatars have fixed places in this room (they are placed close to the corners). Because of the ability to walk around freely, the close-to-real appearance of the avatars located in the IVE, and the immediate and realistic reactions to the participants’ movements, the immersive virtual environment creates a strong feeling of being immersed in the virtual world (Rinck et al., 2010). The virtual environment contained four different avatars (i.e. one in each corner of the room): two baby-related avatars (baby alone, baby–father interaction, see Supplement) and two avatars not related to babies (moving toy, fitness woman, see Supplement). All avatars were moving in a fairly recurrent movement in a limited small area. The participants were given a period of two and a half minutes to walk around as they liked. The virtual walk stopped automatically at the end of this time. Each week the avatars were varied, namely a different moving toy (boat, train, or car), a different interaction between father and child (father bathing, feeding, or changing diaper), baby playing with a different toy (shape sorter, building blocks, or blocks tower), and fitness woman performing a different exercise (arm, abdominal, or upper leg). Two measures were derived from the IVE, namely the mean distance to each avatar and the total time investment. Both these variables represent the behavioral measure of approach (Dotsch & Wigboldus, 2008). The computer records which avatar the participant is looking at by recording the orientation and position of the head. The mean distance to each avatar reflects the average distance from the sensor on top of the participant’s shutter glasses to the center of the avatar during the time the participant was looking at the avatar during the experiment. In a second step, the mean distance was calculated combined for the two baby-related avatars and for the two non-baby-related avatars. The total time investment reflects the total sum of time during which the participant was looking at the avatar during the experiment. Here again, the total time investment was calculated over the two baby-related avatars and over the two non-baby-related avatars. 2.5. Statistical procedure From our total sample of 66 participants (46 fathers-to-be and 20 controls), one father-to-be had missing data for the explicit caregiving interest measure. For the basal hormones, 11 fathers-to-be and three controls missed one or two hormone samples. With respect to the IVE data, 28 participants (16 fathers-to-be and 12 controls) had complete IVE data, 13 participants missed data from one week, and 25 participants missed data from two weeks; these missing values were due to discontinuation of seven participants and technical problems with the virtual reality system. SPSS missing value analysis showed that the IVE values were missing completely at random (MCAR; Little & Rubin, 1987; Little’s MCAR test: Chi-Square = 85.047, df = 72, p = .140). Expectation-maximization (EM) algorithm was used to impute missing values in the IVE dataset, as described by Dempster, Laird, and Rubin (1977). The IVE values higher or lower than 3 SD from the total group mean per condition were regarded as outliers (less than 0.5%) and winsorized, i.e. replaced by the next highest or lowest value (Tukey, 1977). The analyses were carried out by first examining mean differences between the fathers-to-be and the controls for their explicit caregiving interests, and their basal oxytocin and vasopressin levels at all three visits, using multivariate analyses of variance. The basal hormone levels were not normally distributed. Therefore, we performed a logarithmic and inversed transformation on the values of vasopressin and oxytocin, respectively. One outlier higher than 3SD from the total mean was found for oxytocin concentrations at the third visit and winsorized (Tukey, 1977). Second, we performed correlation analyses to test whether basal oxytocin and vasopressin levels were associated with explicit and implicit caregiving interest in the fathers-to-be and the controls. Third, group differences in implicit caregiving interest were tested with the IVE experiment, after the administration of the intranasal sprays. Two 3 × 2 × 2 (‘spray condition’ (placebo, oxytocin and vasopressin) × ‘avatar’ (baby-related vs. nonbaby-related) × ‘group’ (fathers-to-be vs. controls)) mixed between-within repeated measures ANOVA’s were conducted (one for total time investment in the avatars and one for mean distance to the avatars). 3. Results 3.1. Explicit caregiving interests Table 2 describes the means for the fathers-to-be and the controls for their explicit caregiving interests. There was a significant difference between fathers-to-be and the controls on explicit care (F(3,61) = 2.92, p = 0.04; Wilks’ Lambda = .87, partial eta squared = .13). Post hoc analyses showed that the fathers-to-be reported higher interest in direct care for babies (M = .12, SD = .6), compared to controls (M = −.25, SD = .4). 3.2. Basal oxytocin and vasopressin concentrations Table 2 shows basal hormone concentrations for the three visits on the morning of each lab visit. Basal levels of oxytocin and vasopressin correlated significantly for the expectant fathers (r = .67, p < .01), and for the controls (r = .53, p < .01). The multivariate test revealed that the basal hormones per visit did not differ between the fathers-to-be and the controls (p’s > 0.30). Table 3 shows the correlations for the both groups between their explicit caregiving interests, implicit caregiving

C.C.C. Cohen-Bendahan et al. / Infant Behavior & Development 41 (2015) 26–37

31

Table 2 Descriptive statistics of explicit care interest domains and the basal oxytocin and vasopressin levels on the day of each lab visit, separated for fathers-to-be and controls. Mfathers-to-be (SD) Explicit care interest Indirect care Direct care Play Week 1 Basal OT Basal AVP Week 2 Basal OT Basal AVP Week 3 Basal OT Basal AVP

Mcontrols (SD)

.02 (.60) .12 (.56) .08 (.84)

.02 (.42) −.25 (.42)* −.18 (.67)

3.21 (8.73) 2.26 (1.02)

2.79 (4.05) 1.58 (.94)

4.43 (2.18) 4.25 (3.24)

4.39 (2.33) 3.46 (2.23)

4.14 (1.99) 2.86 (1.87)

4.36 (2.11) 2.65 (1.7)

Note: OT: oxytocin (ng/mmol); AVP: vasopressin (ng/mmol). * p < .05. Table 3 Pearson correlations for basal oxytocin and vasopressin concentrations, explicit and implicit caregiving interest, separated for fathers-to-be and controls. Top half (gray) represents the correlations for the controls and lower half for the fathers-to-be. 1 Basal hormone concentrations Basal level OT 1 Basal level AVP 2 Explicit caregiving interest Direct care 3 Indirect care 4 Play 5 Implicit caregiving interest (in placebo condition) Total time baby-related avatars 6 7 Total time non-baby-related avatars Distance baby-related avatars 8 Distance non-baby-related avatars 9

2 .53**

.67** −.21 −.15 −.19

−.09 −.26 −.00

.20 .10 −.07 .17

.08 .07 −.14 .13

3

4

−.14 −.18

−.06 −.14 .18

.51** .69** −.07 .06 −.01 −.12

5

6 .05 .07

.63** −.08

.40** .06 −.02 .13 .15

−.04 −.03 .02 .05

7

8

9

.24 .15

−.40 −.10

−.04 −.17

.35 −.11

.16 −.06 .14

−.14 .01 −.16

.07 .08 .05

.02 .28 −.11

−.45* −.05 −.32* .24

.20 −.11

−.26 .06

−.09 −.29 .75**

.37*

Note: *p < .05, **p < .01; OT: oxytocin; AVP: vasopressin.

interests during the placebo condition and the corresponding basal oxytocin and vasopressin levels. No significant correlations were found between the basal oxytocin and vasopressin levels and explicit and implicit caregiving interest variables in fathers-to-be and controls. Moreover, implicit and explicit caregiving interest were shown not to be correlated. 3.3. Implicit caregiving interests and exogenous hormones Table 4 presents the descriptive statistics for the IVE variables per experimental condition, separately for the fathers-to-be and the controls. 3.3.1. Total time investment The mixed between-within repeated measures ANOVA for the outcome measure “total time investment” in baby-related versus non-baby-related avatars, showed a significant main effect for ‘avatar’ (F(1,64) = 62.13, p < .001, Wilks’ Lambda = .51, partial eta squared = .49). All the participants spent more time looking at the baby-related avatars than at the nonbaby-related avatars (see Table 4). In addition, a significant interaction was found between ‘spray condition’ and ‘group’ Table 4 Mean (SD) of the implicit caregiving interest variables for each spray condition separate for fathers-to-be and controls. Fathers-to-be

Total time (s) Baby-related avatars Non-baby-related avatars Proximity (m) Baby-related avatars Non-baby-related avatars

Controls

Placebo

OT

AVP

Placebo

OT

AVP

62.64 (11.6) 41.79 (10.1)

65.45 (16.0) 42.79 (10.1)

68.8 (22.0) 42.91 (13.9)

61.27 (11.1) 46.34 (10.8)

60.14 (8.8) 45.54 (10.6)

58.2 (11.2) 44.67 (11.9)

1.59 (.4) 1.88 (.3)

1.53 (.3) 1.87 (.3)

1.51 (.3) 1.76 (.3)

1.58 (.4) 1.72 (.3)

1.62 (.3) 1.83 (.3)

Note: OT: oxytocin; AVP: synthetic vasopressin; Total time is displayed in seconds (sum); distance is displayed in meters (mean).

1.61 (.3) 1.74 (.3)

32

C.C.C. Cohen-Bendahan et al. / Infant Behavior & Development 41 (2015) 26–37

Fig. 1. Total time investment and standard error bars in the baby-related avatars during each nasal spray condition. Post hoc analyses revealed the difference in total time looking at the baby-related avatars during the vasopressin (AVP) condition was significantly longer for the fathers-to-be than the controls (*p < .05).

(F(2,63) = 3.14; p < .05, Wilks’ Lambda = .91, partial eta squared = .09), indicating that the groups reacted differently to the hormonal sprays. The fathers-to-be invest more time in the baby and non-baby-related avatars while sprayed by the hormones oxytocin and vasopressin, compared to the placebo condition. In contrast, the controls spent less time in looking at baby-related avatars, and therefore seem to wander more among the avatars, during the two hormonal conditions, compared to the fathers-to-be. Simple main effects post hoc analyses revealed that the difference in total time looking at the baby-related avatars during the vasopressin condition was significantly longer for the fathers-to-be compared to the controls (F(1,64) = 4.127; p < .05, see also Fig. 1). Also, a marginally significant interaction was found between ‘avatar’ and ‘group’ (F(1,64) = 3.41; p = .07, Wilk’s Lambda = .95, partial eta squared = .05). Fathers-to-be tend to spend more time looking at the baby-related avatars (see Fig. 2) compared to the controls. None of the other main effects or interactions were significant.1 3.3.2. Mean distance The mixed between-within repeated measures ANOVA for the outcome measure mean distance toward baby-related versus non-baby-related avatars, showed a significant main effect for ‘avatar’ (F(1,64) = 71.29, p < .001, Wilks’ Lambda = .47, partial eta squared = .53). The participants stand closer to the baby-related avatars, compared to the non-baby-related avatars. Also, there was a significant interaction between ‘avatar’ and ‘group’ (F(1,64) = 6.03, p < .05, Wilks’ Lambda = .91, partial eta squared = .09). This interaction can be found in Fig. 3, and shows that fathers-to-be stand closer to the baby-related avatars, and further away from the non-baby-related avatars, compared to the controls. None of the other main effects or interactions were significant.1 4. Discussion In this study, we examined caregiving interest in men during the third trimester of their partner’s pregnancy (i.e. 32–34 weeks of gestation). Both explicit and implicit measures of caregiving interest were used, and we examined whether the expectant fathers’ caregiving interest was affected by endogenous and exogenous oxytocin and vasopressin. Using the immersive virtual environment (IVE), expectant fathers stood closer to and tended to spend more time looking at the

1 Repeating the analyses with the non-imputed IVE data produces comparable results. The significant and marginally significant findings remain the same.

C.C.C. Cohen-Bendahan et al. / Infant Behavior & Development 41 (2015) 26–37

33

Fig. 2. Total time investment and standard error bars for the non-baby and baby-related avatars. The marginally significant interaction indicates that fathers-to-be tend to spend more time looking at the baby-related avatars, and less time looking at the non-baby-related avatars, than the controls (p = .07).

Fig. 3. Mean distance and standard error bars for the non-baby and baby-related avatars. The analyses show a significant interaction with the fathers-to-be standing closer to the baby-related avatars, and further away from the non-baby-related avatars, compared to the controls (p < .05).

baby-related avatars, and stood further away and tended to spend less time looking at the non-baby-related avatars, compared to the childless control men. In addition, fathers-to-be showed more interest in direct care for children as compared to the control group. Basal oxytocin and vasopressin concentrations were not related to explicit and implicit caregiving interest in expectant fathers, and were not different from those of the childless control men. Finally, when exogenous vasopressin was administered through a nasal spray during the IVE, expectant fathers invest more time watching the baby-related avatars compared to control men.

34

C.C.C. Cohen-Bendahan et al. / Infant Behavior & Development 41 (2015) 26–37

As hypothesized, expectant fathers during the third trimester of their partner’s pregnancy showed more interest in infant caregiving, both explicitly and implicitly. Compared to childless controls, expectant fathers reported more interest in direct care for infants, including changing diapers and feeding the infant. On the other hand, expectant fathers and controls were equally interested in indirect care for an infant, such as shopping for infant’s clothes, and play activities. How can this increased paternal interest in direct care, but not in indirect care and play, be explained? During pregnancy, men consciously consider the type of father they want to become and the style of parenting they want to adopt (Chin, Hall, & Daiches, 2011). Although men seem to be more sure of their long-term approach to fatherhood, they are less certain with regard to, and probably more occupied with, their immediate role as a caregiver and accompanied activities shortly after birth (Deave & Johnson, 2008; St John, Cameron, & McVeigh, 2005). This more immediate focus might possibly explain the increased interest in direct care, but not in play, as this might not yet be in order during the first weeks after birth. Also, a distinctive paternal style and a maternal style are evident, even in families with fathers who are highly involved in care for children (Chin et al., 2011). This possibly explains our results with respect to fathers’ increased interest in direct care for infants, but not increased interest in indirect care, as this latter task might be seen as more of a maternal matter. In sum, fathers-to-be report more interest in activities involving direct care for an infant, but not (yet) in other caregiving activities, possibly reflecting worries about being able to perform direct infant care or reflecting certain attitudes toward the specific caregiving tasks that belong to the fatherhood role. In addition to the reported interest for infant caregiving, fathers-to-be also showed more implicit interest in infants and the paternal role. In the immersive virtual environment (IVE), fathers-to-be stood closer to the baby-related avatars, and stood further away from the non-baby-related avatars, compared to the childless controls. In addition, fathers-to-be tended to spend more time looking at the baby-related avatars, and tended to spend less time looking at the non-babyrelated avatars, compared to the childless controls. These results might argue that expectant fathers not only prepare and adjust for upcoming fatherhood consciously, but also at an automatic and unconscious level. Interestingly, fathers reported interest in infant caregiving was not related to their implicit behavioral measures of caregiving interest. So, fathers who reported more interest in infant caregiving were not always the ones standing closer to and spending more time looking at the baby-related avatars. The question remains how these explicit and implicit measures of fathers’ prenatal interest in caregiving differentially predict the actual amount, type and quality of their caregiving behavior with their own child after birth. Unexpectedly, endogenous and exogenous oxytocin did not appear to be related to expectant fathers’ explicit and implicit caregiving interest, and basal urinary concentrations were not different from those of the childless controls. These latter results resemble the findings from Gubernick, Winslow, Jensen, Jeanotte, and Bowen (1995) in which plasma levels of oxytocin were measured in expectant California biparental mice, as well as current fathers and virgin males. Oxytocin levels were reported to be higher in expectant fathers the day following mating, with levels declining by the twentieth day of gestation and remaining comparable to the other male groups till birth. Also, Gray et al. (2007) investigated urinary oxytocin concentrations in three groups of Jamaican men: fathers living together with their family, fathers visiting, but not living together with their family, and a group of single control men. In line with our results and those of the mice study, no differences were found between the controls and fathers in their oxytocin concentrations. The explanation for the oxytocin results may be that oxytocin is dominantly related to maternal caregiving behavior, and less directly involved in the initiation of paternal care (Gubernick et al., 1995). This explanation is in line with the findings from Taylor et al. (2010) showing that distressed pair-bond relationships are unrelated to oxytocin in men in contrary to women (Taylor et al., 2010). It is also possible that paternal basal oxytocin is not related to the amount of interest in paternal caregiving, but to the style and quality of actual paternal caregiving. Support for this hypothesis is found in studies positively relating paternal baseline oxytocin concentrations with stimulatory play (Feldman et al., 2010), and experimental studies showing that intranasal oxytocin administration elevated the quality of paternal play, including more sensitivity, more positive affect and less hostility (Naber, van IJzendoorn, Deschamps, van Engeland, & Bakermans-Kranenburg, 2010; Naber, Poslawsky, van IJzendoorn, Van Engeland, & Bakermans-Kranenburg, 2013; Weisman, Zagoory-Sharon, & Feldman, 2014). Whether expectant fathers’ oxytocin concentrations are related to their style and quality of paternal caregiving before their partners give birth cannot be determined from our data. Future research could help unravel this question, for instance by investigating the interactions of fathers-to-be with another infant or an infant model (i.e. as in an interactive doll, van Anders, Tolman, & Volling, 2012). For vasopressin, the role in paternal caregiving behavior seems to be more complex. Basal vasopressin concentrations were not related to caregiving interest in expectant fathers, and not different from those of childless controls. Gray et al. (2007) also did not find vasopressin differences between Jamaican fathers and childless controls. However, within the group of fathers, vasopressin concentrations were negatively correlated with the age of the youngest child (≤4 years). Hence, vasopressin concentrations were the highest when a father’s partner had just given birth to an infant and were lower with the child’s increasing age. It is therefore possible that fathers’ vasopressin concentrations do not yet increase during pregnancy, but do so immediately after birth as a result of early father–infant interactions. These increases in vasopressin during early interactions might stimulate brain areas that aid in father-infant bonding (Gray et al., 2007). This hypothesis is supported by a study investigating the neurobiology of mothers’ and fathers’ brain response to own-infant compared to standard-infant videos (Atzil, Hendler, Zagoory-Sharon, Winetraub, & Feldman, 2012). Vasopressin-brain correlations emerged in the inferior-frontal gyrus (IFG) and insula, but only for fathers, and not for mothers (Atzil et al., 2012). The IFG and insula embody functions that are crucial for human social bonding, especially between parents and infants. These

C.C.C. Cohen-Bendahan et al. / Infant Behavior & Development 41 (2015) 26–37

35

findings support a unique role of vasopressin in father–infant bonding in the postpartum period, but this hypothesis needs further testing. When vasopressin was administered exogenously through a nasal spray before the IVE, fathers-to-be invested more time watching the baby-related avatars, as compared to the controls. In animal models, vasopressin injections into the lateral septum from sexually inexperienced prairie vole males elicited paternal behavior, e.g., they spent more time contacting and crouching over pups (De Vries, Wang, & Ferris, 1994). So, while vasopressin increased animal paternal behavior, even in virgin males, in our study this was only true for the expectant fathers, and not the controls. Apparently, at least in humans, exogenous vasopressin can have specific behavioral effects depending on the characteristics of the participant. Thompson, George, Walton, Orr, and Benson (2006) already showed sex-specific effects of exogenous vasopressin, promoting agonistic and affiliative types of responses toward same-sex faces in men and women, respectively. Possibly, vasopressin can also have specific effects within males, as a result of the stage of life they are living in (i.e. without children or transitioning to fatherhood). Although this hypothesis clearly needs testing, the results underline the possible role of vasopressin in paternal behavior, and give some insight into possible causal relations, as administering vasopressin led to more time investment of expectant fathers in baby-related avatars. How exogenous administration of vasopressin might affect postpartum paternal caregiving behavior and paternal-infant bonding, should be investigated in further research, as these findings might also have clinical implications. A strength of this study is the intranasal administration of oxytocin and vasopressin in a double-blind, within-subject, counterbalanced crossover design. Also, to our knowledge, this study is the first to investigate the transition to fatherhood with implicit measures. Using the Immersive Virtual Environment (IVE), we were able to standardize the avatars and to unobtrusively and accurately assess implicit behavioral measures (i.e. the looking time investment and the average distance to the baby-related and the non-baby-related avatars) (Rinck et al., 2010). The fact that our study found that expectant fathers showed more implicit interest in infants and the paternal role, irrespective of their explicit interest in caregiving, sets ground for a new area of research approaching the transition to parenthood and parenting with implicit measures. As stated before, an important following step is to investigate the links between these implicit measures and actual postpartum caregiving behavior. Next to the strengths, limitations of this study should also be noted. First of all, the sample size was relatively small and the group of fathers-to-be fairly homogeneous. This compromises the generalizability of the study. In addition, the group of control men was also small and on average seven years younger than the group of expectant fathers. While the age of the participants was not associated with any of the outcome measures (i.e. explicit and implicit caregiving interest and basal hormone concentrations), future studies should try to recruit an age-matched control group. Second, in the absence of minimally invasive techniques for measuring central hormone levels in humans, urine oxytocin and vasopressin were measured. Central and peripheral hormone levels are not always correlated (Churchland & Winkielman, 2012; Landgraf & Neumann, 2004), although there is emerging evidence supporting the correlation between central and peripheral concentrations (Rilling et al., 2012; Ross et al., 2009). However, the possibility exists that central oxytocin and vasopressin concentrations, in contrast to the urine concentrations measured in the current study, are related to fathers’ explicit and implicit caregiving interest. For example, a recent study showed that oxytocin in plasma and saliva, were uncorrelated with oxytocin in urine (Feldman et al., 2011). Nevertheless, we used the first morning void urine while Feldman et al. (2011) used urine collected shortly before and after the experiment. We believe that for basal hormone assessment, the collection of first morning void is to be preferred over urine collection during the day, because the first morning void is less affected by factors such as human interaction and physical activity, and therefore the variability that is caused by these factors is reduced (Witte et al., 2009). Moreover, urinary collection of several hours may reflect more the tonic responses of the participants to their overall social state and less the phasic responses (Moscovice & Ziegler, 2012). Third, there was no check of the intranasal hormone administration to ensure that the manipulation was successful. Future studies should incorporate a manipulation check and collect preferably serum vasopressin and oxytocin samples after the administration to see if the hormone concentrations are raised. Finally, our participants were acquainted with the purpose of the study, i.e. to investigate expectant fathers during their transition to parenthood, and their group status. Response demands could therefore be responsible for our findings regarding implicit and explicit caregiving interest. Although it would be better to recruit a more naïve group of participants, it would be ethically irresponsible to withhold information about the study’s purpose. In sum, expectant fathers during the third trimester of their partner’s pregnancy show more interest in infants and caregiving, both explicitly and implicitly, as compared to childless controls in stable heterosexual relationships. These results might argue that expectant fathers not only prepare and adjust for upcoming fatherhood consciously, but also at an automatic and unconscious level. Moreover, endogenous and exogenous oxytocin did not appear to be related to expectant fathers’ caregiving interest, suggesting that this hormone might not have a prominent role in the initiation of paternal care during pregnancy. The role of vasopressin in paternal caregiving interest seemed to be more complex. Basal vasopressin concentrations were not related to caregiving interest in expectant fathers, and not different from those of childless controls. Nevertheless, when exogenous vasopressin was administered, expectant fathers spent more time watching the baby-related avatars, as compared to control men. These results point in the direction of a role of vasopressin in triggering expectant fathers’ interest in infants and the paternal role. Perhaps vasopressin is more involved than oxytocin in the transition to fatherhood while oxytocin, as Feldman et al. (2010) have shown, is more involved in the actual paternal caregiving behavior. However, more research is needed to unravel the neurobiological mechanisms further and to relate the effects of vasopressin nasal spray to actual paternal caregiving behavior and father–infant bonding.

36

C.C.C. Cohen-Bendahan et al. / Infant Behavior & Development 41 (2015) 26–37

Funding source This research was supported by a personal Veni grant to the first author (Netherlands Organization for Scientific Research, NWO, grant number: 451-07-003). Conflict of interest There were no conflicts of interest, financial or otherwise to declare. Acknowledgements We would like to thank all the men that participated in our study. In addition Jeroen Derks, Hubert Voogd and the RIVERlab team, we are grateful for their technical support during the experiments, and Marlous van Zuijlen, Meike de Vries and Hans van Pelt for their assistance. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.infbeh.2015.06.007. References Atzil, S., Hendler, T., Zagoory-Sharon, O., Winetraub, Y., & Feldman, R. (2012). Synchrony and specificity in the maternal and the paternal brain: Relations to oxytocin and vasopressin. Journal of the American Academy of Child & Adolescent Psychiatry: 51., (8), 798–811. Berg, S. J., & Wynne-Edwards, K. E. (2001). Changes in testosterone, cortisol, and estradiol levels in men becoming fathers. Mayo Foundation for Medical Education and Research: 76., (6), 582–592. Born, J., Lange, T., Kern, W., McGregor, G. P., Bickel, U., & Fehm, H. L. (2002). Sniffing neuropeptides: A transnasal approach to the human brain. Nature Neuroscience: 5., (6), 514–516. Born, J., Pietrowsky, R., & Fehm, H. L. (1999). Neuropsychological effects of vasopressin in healthy humans. Progress in Brain Research: 119., 619–642. Bosch, O. J., & Neumann, I. D. (2012). Both oxytocin and vasopressin are mediators of maternal care and aggression in rodents: From central release to sites of action. Hormones and Behavior: 61., (3), 293–303. Bridges, R. (2010). Neurobiology of the parental brain. Access Online via Elsevier. Brown, R. E. (1993). Hormonal and experiential factors influencing parental behavior in male rodents: An integrative approach. Behavioural Processes: 30., (1), 1–27. Cabrera, N., Tamis-LeMonda, C. S., Bradley, R. H., Hofferth, S., & Lamb, M. E. (2000). Fatherhood in the twenty-first century. Child Development: 71., (1), 127–136. Carter, C. S. (1998). Neuroendocrine perspectives on social attachment and love. Psychoneuroendocrinology: 23., (8), 779–818. Carter, C. S. (2007). Sex differences in oxytocin and vasopressin: Implications for autism spectrum disorders? Behavioural Brain Research: 176., 170–186. Chin, R., Hall, P., & Daiches, A. (2011). Fathers’ experiences of their transition to fatherhood: A metasynthesis. Journal of Reproductive and Infant Psychology: 29., (1), 4–18. Churchland, P. S., & Winkielman, P. (2012). Modulating social behavior with oxytocin: How does it work? What does it mean? Hormones and Behavior: 61., (3), 392–399. Cohen-Bendahan, C. C. C., van Doornen, L. J. P., & de Weerth, C. (2014). Young adults’ reactions to infant crying. Infant Behavior and Development: 37., 33–43. Deave, T., & Johnson, D. (2008). The transition to parenthood: What does it mean for fathers? Journal of Advanced Nursing: 63., (6), 626–633. Dempster, A. P., Laird, N. M., & Rubin, D. B. (1977). Maximum likelihood from incomplete data via the EM algorithm. Journal of the Royal Statistical Society. Series B (Methodological), 1–38. De Vries, G., Wang, Z. X., & Ferris, C. F. (1994). The role of septal vasopressin innervation in paternal behavior in prairie voles (Microtus ochrogaster). Proceedings of the National Academy of Sciences U S A: 91., 400–404. Dotsch, R., & Wigboldus, D. H. (2008). Virtual prejudice. Journal of Experimental Social Psychology: 44., (4), 1194–1198. Feldman, R., Gordon, I., Schneiderman, I., Weisman, O., & Zagoory-Sharon, O. (2010). Natural variations in maternal and paternal care are associated with systematic changes in oxytocin following parent–infant contact. Psychoneuroendocrinology: 35., (8), 1133–1141. Feldman, R., Gordon, I., & Zagoory-Sharon, O. (2011). Maternal and paternal plasma, salivary, and urinary oxytocin and parent–infant synchrony: Considering stress and affiliation components of human bonding. Developmental Science: 14., 752–761. Fernandez-Duque, E., Valeggia, C. R., & Mendoza, S. P. (2009). The biology of paternal care in human and nonhuman primates. Annual Review of Anthropology: 38., 115–130. Gettler, L. T., McDade, T. W., Feranil, A. B., & Kuzawa, C. W. (2011). Longitudinal evidence that fatherhood decreases testosterone in human males. Proceedings of the National Academy of Sciences: 108., (39), 16194–16199. Gordon, I., Zagoory-Sharon, O., Leckman, J. F., & Feldman, R. (2010). Oxytocin and the development of parenting in humans. Biological Psychiatry: 68., (4), 377–382. Gray, P. B., Parkin, J. C., & Samms-Vaughan, M. E. (2007). Hormonal correlates of human paternal interactions: A hospital-based investigation in urban Jamaica. Hormones and Behavior: 52., (4), 499–507. Gubernick, D. J., Winslow, J. T., Jensen, P., Jeanotte, L., & Bowen, J. (1995). Oxytocin changes in males over the reproductive cycle in the monogamous, biparental California mouse, Peromyscus californicus. Hormones and Behavior: 29., (1), 59–73. Heinrichs, M., & Domes, G. (2008). Neuropeptides and social behaviour: Effects of oxytocin and vasopressin in humans. Progress in Brain Research: 170., 337–350. Heinrichs, M., & Gaab, J. (2007). Neuroendocrine mechanisms of stress and social interaction: Implications for mental disorders. Current Opinion in Psychiatry: 20., (2), 158–162. Hewlett, B. S. (1991). Demography and childcare in preindustrial societies. Journal of Anthropological Research, 1–37. Hirschenhauser, K., Frigerio, D., Grammer, K., & Magnusson, M. S. (2002). Monthly patterns of testosterone and behavior in prospective fathers. Hormones and Behavior: 42., (2), 172–181. Kentner, A. C., Abizaid, A., & Bielajew, C. (2010). Modeling dad: Animal models of paternal behavior. Neuroscience & Biobehavioral Reviews: 34., (3), 438–451. Landgraf, R., & Neumann, I. D. (2004). Vasopressin and oxytocin release within the brain: A dynamic concept of multiple and variable modes of neuropeptide communication. Frontiers in Neuroendocrinology: 25., (3), 150–176.

C.C.C. Cohen-Bendahan et al. / Infant Behavior & Development 41 (2015) 26–37

37

Lansu, T. A., Cillessen, A. H., & Karremans, J. C. (2012). Implicit associations with popularity in early adolescence: An approach–avoidance analysis. Developmental Psychology: 48., (1), 65. Lamb, M. E. (Ed.). (2010). The role of the father in child development. Wiley. Levine, A., Zagoory-Sharon, O., Feldman, R., & Weller, A. (2007). Oxytocin during pregnancy and early postpartum: Individual patterns and maternal–fetal attachment. Peptides: 28., (6), 1162–1169. Little, R. J., & Rubin, D. B. (1987). . Statistical analysis with missing data : Vol. 539. (Vol. 539) New York: Wiley. MacDonald, K., & MacDonald, T. M. (2010). The peptide that binds: A systematic review of oxytocin and its prosocial effects in humans. Harvard Review of Psychiatry: 18., (1), 1–21. Marlowe, F. (2000). Paternal investment and the human mating system. Behavioural Processes: 51., (1), 45–61. Moscovice, L. R., & Ziegler, T. E. (2012). Peripheral oxytocin in female baboons relates to estrous state and maintenance of sexual consortships. Hormones and Behavior: 62., (5), 592–597. Naber, F., van IJzendoorn, M. H., Deschamps, P., van Engeland, H., & Bakermans-Kranenburg, M. J. (2010). Intranasal oxytocin increases fathers’ observed responsiveness during play with their children: A double-blind within-subject experiment. Psychoneuroendocrinology: 35., (10), 1583–1586. Naber, F. B., Poslawsky, I. E., van IJzendoorn, M. H., Van Engeland, H., & Bakermans-Kranenburg, M. J. (2013). Brief report: Oxytocin enhances paternal sensitivity to a child with autism: A double-blind within-subject experiment with intranasally administered oxytocin. Journal of Autism and Developmental Disorders: 43., (1), 224–229. Neumann, I. D., Maloumby, R., Beiderbeck, D. I., Lukas, M., & Landgraf, R. (2013). Increased brain and plasma oxytocin after nasal and peripheral administration in rats and mice. Psychoneuroendocrinology: 38., (10), 1985–1993. Paulhus, D. L. (1991). Measurement and control of response bias. San Diego, CA, US: Academic Press. Pietrowsky, R., Strüben, C., Mölle, M., Fehm, H. L., & Born, J. (1996). Brain potential changes after intranasal vs. intravenous administration of vasopressin: Evidence for a direct nose-brain pathway for peptide effects in humans. Biological Psychiatry: 39., (5), 332–340. Reburn, C. J., & Wynne-Edwards, K. E. (1999). Hormonal changes in males of a naturally biparental and a uniparental mammal. Hormones and Behavior: 35., (2), 163–176. Rilling, J. K. (2013). The neural and hormonal bases of human parental care. Neuropsychologia: 51., (4), 731–747. Rilling, J. K., DeMarco, A. C., Hackett, P. D., Thompson, R., Ditzen, B., Patel, R., et al. (2012). Effects of intranasal oxytocin and vasopressin on cooperative behavior and associated brain activity in men. Psychoneuroendocrinology: 37., (4), 447–461. Rinck, M., Kwakkenbos, L., Dotsch, R., Wigboldus, D. H., & Becker, E. S. (2010). Attentional and behavioural responses of spider fearfuls to virtual spiders. Cognition and Emotion: 24., (7), 1199–1206. Ross, H. E., Cole, C. D., Smith, Y., Neumann, I. D., Landgraf, R., Murphy, A. Z., et al. (2009). Characterization of the oxytocin system regulating affiliative behavior in female prairie voles. Neuroscience: 162., (4), 892–903. Sarkadi, A., Kristiansson, R., Oberklaid, F., & Bremberg, S. (2008). Fathers’ involvement and children’s developmental outcomes: A systematic review of longitudinal studies. Acta Paediatrica: 97., (2), 153–158. St John, W., Cameron, C., & McVeigh, C. (2005). Meeting the challenge of new fatherhood during the early weeks. Journal of Obstetric, Gynecologic & Neonatal Nursing: 34., (2), 180–189. Storey, A. E., Walsh, C. J., Quinton, R. L., & Wynne-Edwards, K. E. (2000). Hormonal correlates of paternal responsiveness in new and expectant fathers. Evolution and Human Behavior: 21., (2), 79–95. Storm, E. E., & Tecott, L. H. (2005). Social circuits: Peptidergic regulation of mammalian social behavior. Neuron: 47., 483–486. Taylor, S. E., Saphire-Bernstein, S., & Seeman, T. E. (2010). Are plasma oxytocin in women and plasma vasopressin in men biomarkers of distressed pair-bond relationships? Psychological Science: 21., (1), 3–7. Thompson, R. R., George, K., Walton, J. C., Orr, S. P., & Benson, J. (2006). Sex-specific influences of vasopressin on human social communication. Proceedings of the National Academy of Sciences: 103., (20), 7889–7894. Tukey, J. W. (1977). Exploratory data analysis. Reading, MA: Addison Wesley. van Anders, S. M., Tolman, R. M., & Volling, B. L. (2012). Baby cries and nurturance affect testosterone in men. Hormones and Behavior: 61., (1), 31–36. Wang, Z. X., Liu, Y., Young, L. J., & Insel, T. R. (2000). Hypothalamic vasopressin gene expression increases in both males and females postpartum in a biparental rodent. Journal of Neuroendocrinology: 12., (2), 111–120. Weisman, O., Zagoory-Sharon, O., & Feldman, R. (2014). Oxytocin administration, salivary testosterone, and father–infant social behavior. Progress in Neuro-Psychopharmacology and Biological Psychiatry: 49., 47–52. Witte, E. C., Heerspink, H. J. L., de Zeeuw, D., Bakker, S. J., de Jong, P. E., & Gansevoort, R. (2009). First morning voids are more reliable than spot urine samples to assess microalbuminuria. Journal of the American Society of Nephrology: 20., (2), 436–443. Wynne-Edward, K. E. (2001). Hormonal changes in mammalian fathers. Hormones and Behavior: 40., 139–145.