ArticlePDF Available

Theory of mind processing in expectant fathers: Associations with prenatal oxytocin and parental attunement

  • National Center for PTSD and Boston University School of Medicine Department of Psychiatry


Social cognition may facilitate fathers' sensitive caregiving behavior. We administered the Why‐How Task, an fMRI task that elicits theory of mind processing, to expectant fathers (n = 39) who also visited the laboratory during their partner's pregnancy and provided a plasma sample for oxytocin assay. Three months postpartum, fathers reported their beliefs about parenting. When rating “Why” an action was being performed versus “How” the action was being performed (Why > How contrast), participants showed activation in regions theorized to support theory of mind, including the dorsomedial prefrontal cortex and superior temporal sulcus. Fathers' prenatal oxytocin levels predicted greater signal change during the Why > How contrast in the inferior parietal lobule. Both prenatal oxytocin and attunement parenting beliefs were associated with Why > How activation in the dorsolateral prefrontal cortex, a theory of mind region implicated in emotion regulation. Posterior parahippocampal gyrus and dorsolateral prefrontal cortex activation during the Why > How contrast predicted fathers' attunement parenting beliefs. In conclusion, fathers' neural activation when engaging in a theory of mind task was associated with their prenatal oxytocin levels and their postpartum attunement parenting beliefs. Results suggest biological and cognitive components of fathering may track with the theory of mind processing.
Developmental Psychobiology. 2021;00:1–19.
1© 2021 Wiley Periodicals LLC
Father caregiving behavior yields long- term benefits to children
in social, behavioral, and cognitive domains (Sarkadi et al., 2008).
Extensive research examines the complex underpinnings of moth-
ers' relationship with infants, including biological, behavioral, and
psychological adaptations that serve to suppor t the infants' survival
(Norholt, 2020). Infants form bonds with fathers (Cabrera et al.,
2018), but less remains known about the neurocognitive mecha-
nisms that support the transition to fatherhood. Within the context
of father– infant relationships, social cognition— specifically, theory
of mind— may support fathers' ability to provide sensitive and at-
tuned caregiving, allowing fathers to infer what infants feel and need
(Abraham et al., 2014; Frith & Frith, 2006). Preliminary research
suggests that the neural regions that support the theory of mind
(e.g., superior temporal sulcus) may also underlie fathering behavior
(Abraham et al., 2014). Fathers' theory of mind ability may be a key
building block underlying fathers' engagement in attuned caregiving
The research literature on the fathering brain is scarce and
preliminary. Most parenting- focused neuroimaging research has
measured parents' neural responses to infant stimuli, such as
Received: 5 August 2020 
Revised: 17 January 2021 
Accepted: 13 February 2021
DOI: 10.1002 /dev.22115
Theory of mind processing in expectant fathers: Associations
with prenatal oxytocin and parental attunement
Sofia I. Cardenas | Sarah A. Stoycos | Pia Sellery |
Narcis Marshall | Hannah Khoddam | Jonas Kaplan | Diane Goldenberg |
Darby E. Saxbe
Department of Psychology, University of
Southern California, Los Angeles, CA , USA
Darby E. S axbe, Department of
Psychology, University of Southern
Califo rnia, Los An geles, CA 90089, USA.
Funding information
Rober t Wood Johnson Foundation;
National Science Foundation, Grant/
Award Number: 1552452 and 1842487
Social cognition may facilitate fathers' sensitive caregiving behavior. We administered
the Why- How Task, an fMRI task that elicits theory of mind processing, to expect-
ant fathers (n = 39) who also visited the laboratory during their partner's pregnancy
and provided a plasma sample for oxytocin assay. Three months postpartum, fathers
reported their beliefs about parenting. When rating “Why” an action was being per-
formed versus “How” the action was being performed (Why > How contrast), partici-
pants showed activation in regions theorized to support theor y of mind, including the
dorsomedial prefrontal cor tex and superior temporal sulcus. Fathers' prenatal oxy-
tocin levels predicted greater signal change during the Why > How contrast in the
inferior parietal lobule. Both prenatal oxy tocin and attunement parenting beliefs were
associated with Why > How activation in the dorsolateral prefrontal cortex, a theor y
of mind region implicated in emotion regulation. Posterior parahippocampal gyrus and
dorsolateral prefrontal cortex activation during the Why > How contrast predicted
fathers' attunement parenting beliefs. In conclusion, fathers' neural activation when
engaging in a theory of mind task was associated with their prenatal oxytocin levels
and their postpartum attunement parenting beliefs. Results suggest biological and
cognitive components of fathering may track with the theory of mind processing.
fatherhood, neuroimaging, oxytocin, parental attunement, theory of mind
photographs, infant cr y sounds, or video clips of the infant (e.g., Atzil
et al., 2012; Khoddam et al., 2020; Kuo et al., 2012). These stud-
ies have found associations between neural activity in brain regions
broadly supporting social cognition, including theory of mind, and
the quality of fathering behaviors. However, these studies do not
specify the mental processes that fathers engage in when watching
infant stimuli, leaving unanswered questions. Thus, although prior
studies provide insight into the neural correlates of engaging with
infant stimuli, there is a gap in understanding the neural correlates
of theory of mind in parents, despite its importance for later par-
enting behavior. Furthermore, though several studies have linked
hormones with the neural systems implicated in fathering (Abraham
et al., 2014; Mascaro et al., 2014; Witt foth- Schardt et al., 2012),
studies that use tasks meant to isolate specific socio- emotional pro-
cesses would allow for the identification of processes implicated in
father neurobiology. Emerging evidence suggests changes in men's
hormones before pregnancy may indicate preparation for caregiving
(Khoddam et al., 2020; Saxbe et al., 2017). However, no studies to
our knowledge have looked specifically at prenatal levels of oxyto-
cin, a neuropeptide hormone that has been associated with social
affiliative behavior, within expectant fathers.
This study addresses these gaps in the literature by investigat-
ing the neural correlates of theory of mind within expectant fa-
thers. We also examined whether paternal prenatal oxytocin levels
predict neural activation to theory of mind and whether prenatal
activation during theory of mind predicts subsequent postpartum
self- reported paternal beliefs about parenting. Fathers specifically
reported on whether they believe infants benefit from parenting
that is more attuned— that is, more sensitive and responsive to their
behaviors and signals.
1.1  |  Fathering behavior and theory of mind
The neurobiology of paternal caregiving is an emerging field of re-
search that overlaps with existing research on the neurobiology of
motherhood. Extensive literature evaluates the neural basis of par-
enting be haviors in human ma ternal caregive rs (Cárdenas et a l., 2019;
Rilling & Mascaro, 2017). One seminal study found that the human
maternal brain undergoes structural changes in gray matter across
the prenatal and postpartum period, including regions involved in
theory of mind and language processing (e.g., superior temporal sul-
cus, middle and superior temporal gyrus, parahippocampal gyrus)
and emotion, memory, and self- awareness (e.g., precuneus, posterior
cingulate cortex); these structural changes predicted parent- related
outcomes, such as mother- to- infant attachment, degree of hostility
toward the infant, and functional ac tivity when looking at own ver-
sus other infants (Hoek zema et al., 2017). Another longitudinal study
looked at structural gray- matter changes in fathers' brains during
the postpartum period (from 2– 4 weeks postpar tum to 12 16 weeks
postpartum; Kim et al., 2014); the study found that fathers had in-
creased gray matter volume in regions underlying motivation and
attachment (e.g., hypothalamus, amygdala, striatum) and decreases
in gray matter volume in regions involved in processing threat and
ambiguit y (e.g., orbitofrontal cortex, and insula) and self- awareness
(e.g., posterior cingulate cortex). Thus, preliminary studies suggest
human maternal and paternal brains show structural changes across
the transition to parenthood that may support caregiving behaviors.
The human maternal brain recruits several subcor tical regions
involved in motivation (Gregory et al., 2015) and cortical systems
involved in cognitive and emotional empathy (Rilling, 2013). One of
the cortical systems supports the abilit y to consider the thoughts
and feelings of others (i.e., theory of mind). This neural “theor y of
mind” system is composed of a group of regions including temporo-
parietal junction, dorsomedial prefrontal cortex, ventromedial pre-
frontal cortex, superior temporal sulcus/superior temporal gyrus,
and precuneus (Gallagher & Frith, 2003; Molenberghs et al., 2016).
An emerging body of research applies neurobiological methods
for understanding paternal caregiving behavior. Atzil et al. (2012)
measured mothers' and fathers' brain activity in response to their
infants' videos. While mothers had greater activity in limbic areas,
such as the right amygdala, fathers had greater activity in social-
cognitive cortical areas, such as the dorsal prefrontal cortex. A
second study by the same research group (Abraham et al., 2014)
presented first- time mothers and fathers in an fMRI scanner with
videos of themselves playing with their infants and a stranger play-
ing with a strange infant. They compared three groups of first- time
parents with varying degrees of involvement in childrearing: primary
caregiver mothers in a heterosexual relationship, secondary care-
giver fathers in a heterosexual relationship, and primary caregiver
fathers in a homosexual relationship raising infants without mater-
nal involvement. When contrasting self- infant interactions and un-
familiar parent– infant interaction, the study found similar levels of
activation in brain regions supporting emotion and motivation (e.g.,
ventral anterior cingulate cortex, ventral tegmental area, anterior
insula/inferior frontal gyrus) and mentalizing network (e.g., ventro-
medial prefrontal cortex, temporal poles, lateral frontopolar cortex).
However, primary caregiver mothers and primary caregiver fathers
showed greater activation in the amygdala to self- infant interactions
versus unfamiliar parentinfant interactions compared to secondary
caregiver fathers. When viewing self- infant interactions versus un-
familiar parent– infant interactions, both groups of fathers exhibited
greater ac tivation in the superior temporal sulcus, an area widely
associated with the theory of mind. Findings from these two studies
suggest that the neurobiology of fatherhood has some overlap and
some distinction from that of mothers and the degree of caregiving
behavior may play a role in how caregivers process infant- related
stimuli. Additionally, these studies provide preliminary evidence that
the theory of mind network may play a par ticularly important role in
supporting fathering behavior.
1.2  |  Measuring neural correlates of theory of mind
Theory of mind is the consideration of the mental states and inten-
tions of people (Weimer et al., 2017). Theory of mind also allows
individuals to perceive the behavior as driven by unobservable
thought s (Spunt & Lieberman, 2012). By age four, most typic ally de-
veloping children achieve an understanding of the mind (Cutting &
Dunn, 1999). However, knowledge about mental states continues to
increase after the age of four, including interpretive theor y of mind
or the ability to understand that individuals interpret events differ-
ently (Carpendale & Chandler, 1996). Differences in theory of mind
abilities for children correlate with higher general social competence
(e.g., end conversations appropriately, engage in cooperative play
behaviors, follow rules during games; Lalonde & Chandler, 1995).
Theory of mind abilities improve in typically developing adults be-
tween late adolescence and adulthood (Dumontheil et al., 2010).
Thus, theory of mind has implications for social competence and can
change across development.
Extensive research seeks to develop accurate behavioral and
neural measures that reflect the multidimensional, dynamic, and
variable nature of the theory of mind construct (e.g., Keysar et al.,
2003). Neuroimaging studies using theory of mind- related tasks
reveal a consistent activation of brain regions including the medial
prefrontal cortex, temporal poles, temporoparietal junction, pre-
cuneus, and posterior superior temporal sulcus (e.g., Frith & Frith,
2003, 20 06; Gallagher et al., 2000; Saxe et al., 2006). However,
activation in the theory of mind net work varies across individual
studies based on the mental state evaluated (belief vs. desire) and
the stimuli used (verbal vs. non- verbal; Spunt & Adolphs, 2014).
This study used the Why- How Task, which isolates the multiple
cognitive processes underlying the day- to- day theory of mind
that occurs when individuals perceive others' actions (Spunt &
Lieberman, 2012). The task is one of several variants created by
a team of researchers studying social cognition (Spunt & Adolphs,
2014; Spunt & Lieberman, 2012). The “Why” process is supported
by brain regions associated with theory of mind, whereas the
“How” process is associated with areas related to action percep-
tion (Spunt & Lieberman, 2012). In the task, participants watch
sixteen blocks of eight photographs. Each of the blocks leads to
a response question meant to elicit processes associated with ac-
tion perception (“Is the person looking sideways?”) or theory of
mind (“Is the person helping?”; Spunt & Adolphs, 2014).
1.3  |  The role of oxytocin
Some research suggests that oxy tocin may support social cognition
and pro- sociality in humans (for review, see Bartz et al., 2011). One
study found that men in a within- subject, double- blind, placebo-
controlled trial performed better in the Reading the Mind in the Eyes
Test (RMET) after receiving the oxytocin administration compared
to men who received placebo (Domes et al., 2007). In a double- blind,
placebo- controlled study, Yue et al. (2017) found that women who
inhaled ox ytocin versus placebo displayed faster response times
when judging others' perspectives. This same ef fect was not found
in males, indicating that oxytocin may differently impact men's and
women's perspective- taking (Yue et al., 2017). Although some data
in men and women suggest that oxytocin may causally support so cial
cognitive processes, participants' sex must be t aken into account.
Many studies have investigated the neuroendocrine under-
pinnings of parenting (Gordon et al., 2010; Scatliffe et al., 2019).
Some studies have found associations between fathers' levels of
hormones, including oxytocin, testosterone, and vasopressin, and
their caregiving behavior (Feldman & Bakermans- Kranenburg, 2017;
Gettler, 2014; Storey & Ziegler, 2016). Oxytocin has been associated
with social and affiliative behavior, including behavioral synchrony
and responsiveness in mothers and fathers (Feldman et al., 2010;
Morris et al., under review; Scatliffe et al., 2019). In fathers, oxytocin
has been specifically implicated in stimulator y parenting behaviors
(Feldman et al., 2010; Scatliffe et al., 2019) supporting fatherinfant
attachment (George et al., 2010). Furthermore, experimentally ad-
ministered oxytocin levels elicit more stimulating play in fathers
(i.e., interactions that promote child exploration, often described as
“rough and tumble” play; Gordon et al., 2010; Naber et al., 2010;
Weisman et al., 2014). However, to our knowledge, no study has
assessed whether prenatal oxytocin levels in expectant fathers are
linked with enhanced postpartum caregiving outcomes or attitudes.
Furthermore, the prior literature does not reconcile whether men's
ability to engage in theory of mind is linked with prenatal oxyto-
cin and subsequent postpartum father caregiving attitudes and
Few studies have investigated associations between oxytocin,
brain activity, and father caregiving behavior (e.g., Abraham et al.,
2014; Atzil et al., 2012). Atzil et al. (2012) had mothers and fathers in
an fMRI scanner watched videos of own infant play and other infant
play. Mothers showed relatively greater activation in subcortical
regions linked with emotion processing (e.g., amygdala) and reward
(e.g., caudate), whereas fathers had greater activation in social cog-
nitive net works that support social cognition and empathy (i.e., me-
dial PFC, precuneus, and inferior parietal gyri). Furthermore, fathers'
postpartum plasma oxytocin levels negatively correlated with activ-
ity in the left inferior and superior frontal gyri, left primary motor
cortex, medial prefrontal cortex, and left anterior cingulate cortex
(Atzil et al., 2012). Building upon this literature, Abraham et al. (2014)
assessed oxytocin, brain activity, and parent c aregiving behavior.
All participants provided salivary oxytocin data, completed a func-
tional MRI task in which they watched videos of themselves playing
with their infant (i.e., self- infant interaction) or an unfamiliar parent
playing with an unfamiliar infant (i.e., unfamiliar parentinfant), and
took part in a father– infant play task micro- coded for parent– infant
behavioral synchrony. Although oxytocin levels were similar across
mothers and fathers, fathers showed greater activity in a region of
the brain involved in theory of mind, perspective- taking, and empa-
thy (i.e., superior temporal sulcus). Fathers' superior temporal sulcus
(STS) activation while watching infant videos correlated positively
with parentinfant synchrony as well as ox ytocin levels. The rela-
tionship between STS and synchrony was mediated by oxytocin.
This study suggests that fathers' social cognition, oxytocin levels,
and fathering behavior may be interrelated constructs. However,
as this study used a non- standardized fMRI task, further research
should clarify how oxytocin interacts with specific social- cognitive
processes. Additionally, this study used salivary oxytocin assayed
without extraction. Although salivary oxytocin measures have been
utilized in the parenting literature, there is some doubt about the
reliability of these measures and uncertainty of the mechanism by
which oxytocin enters the saliva (for review, see MacLean et al.,
Surprisingly, no studies to our knowledge have assessed prena-
tal oxy tocin levels in expectant fathers. Men and women have been
found to show shifts in hormone levels from the prenatal to the
postpartum period, and some evidence suggests that these shifts in
hormones support parenting outcomes (Edelstein et al., 2017; Saxbe
et al., 2017). For instance, Storey et al. (2000) found that men and
women had similar stage- specific shifts in cortisol, testosterone,
and prolactin across the prenatal and postpar tum periods. However,
some research suggests that oxy tocin does not shift similarly across
pregnancy. An older study by Leake et al. (1981) found that healthy
men, non- pregnant women, and pregnant women (15– 42 weeks
gestation) had similar levels of plasma oxy tocin, suggesting that oxy-
tocin levels during pregnancy may not change. In contrast, Gordon
et al. (2010) found intraindividual stability during the first 6 months
postpartum and interrelated plasma oxytocin levels in mother- father
dyads at 1 and 6 months postpartum. Fur thermore, paternal oxy-
tocin correlated with stimulatory parenting behavior. These find-
ings suggest that oxy tocin levels may be consistent across time and
provide preliminary support for investigating the utility of prena-
tal oxy tocin in men as a predictor of postpartum father caregiving
Theory of mind processing may inform paternal beliefs and cog-
nitions about parenting. Parents' tendency to consider the thoughts
and need s of infants (i.e., en gage in theor y of mind) is related to tim ely
and appropriate responsiveness to infants' needs (Ereky- Stevens,
2008). Prior research suggests that parents' beliefs and principles
about parenting are associated with their actual parenting behaviors
during early infancy (Winstanley & Gattis, 2013). However, to our
knowledge, no studies have assessed whether expec tant fathers'
theory of mind activation relates to any aspects of their postpartum
parenting, including their caregiving beliefs and cognitions.
1.4  |  The current study
Although neural and behavioral research suggests that oxytocin and
theory of mind may support fathers in the transition to parenthood,
our paper uniquely uses a longitudinal approach that combines a
standardized prenatal fMRI theory of mind task, prenatal ox ytocin
levels, and postpartum beliefs and cognitions about parenting. This
study used the Why- How Task (Spunt & Adolphs, 2014), an fMRI
task designed to elicit theory of mind processing within a sample
of expectant fathers. Moreover, we used immunoassay with extrac-
tion to measure oxytocin levels in plasma, which is recommended
for plasma oxytocin assay (Lefevre et al., 2017). We controlled for
fathers' education, given evidence that education may be associated
with fathering behavior (Cabrera et al., 2007; Malin et al., 2012). We
also controlled for partners' pregnancy stage.
We tested three hypotheses:
(i) Consistent with Spunt and Adolphs (2014), we predicted that ex-
pectant fathers would show greater signal change during the
Why versus How contrast in regions that have been associated
with theory of mind specifically (e.g., dorsomedial prefrontal cor-
tex; ventromedial prefrontal cortex; orbitofrontal cortex; tem-
poroparietal junction; posterior cingulate cortex; temporal pole;
anterior superior temporal sulcus).
(ii) As discussed earlier, oxytocin may influence fathering behavior
and may be associated with neural activity in regions linked to
theory of mind (Abraham et al., 2014). We hypothesized that
fathers with higher prenatal oxy tocin levels would show more
signal change in brain regions associated with theory of mind
processing (i.e., brain regions listed in hypothesis i) during the
Why > How contrast.
(iii) We hypothesized that activation during the Why > How contrast
in theor y of mind regions would be associated with greater en-
dorsement of thoughts and beliefs associated with attuned care-
giving behavior.
2.1  |  Participants
This study uses data from the larger longitudinal Hormones and
Attachment across the Transition to Childrearing (HATCh) study,
which follows couples from the mothers' mid- to- late pregnancy
through the first year postpartum. We recruited participants
through social media advertising (e.g., Facebook, online parenting
groups), flyers, and word of mouth. This study used self- reported
data provided by fathers from prenatal laborator y visits, con-
ducted with both members of the couple when mothers were in
mid- to- late pregnancy, subsequent father- only MRI visits that
we scheduled within approximately 2 weeks of the prenatal in-
laboratory visit, and postpartum questionnaires collected online
approximately 3 months after the birth. Eligible couples were co-
habitating, first- time parents of a singleton pregnancy, and did not
report any medications or conditions known to interfere with en-
docrine system hormones (e.g., steroid medicines, Cushing's dis-
ease). We also excluded couples who reported psychiatric illness
requiring medication or illegal drug use. Users of tobacco, mari-
juana, and we allowed some psychiatric medications if able to ab-
stain for 24 h before their study visit. Additional exclusion criteria
included contraindications for magnetic resonance imaging (MRI),
such as left- handedness, neurological or movement disorders,
claustrophobia, history of brain injury, psychotropic medication,
or severe learning disability. Additionally, participants needed
to have sufficient English language fluency to complete study
measures and scanning procedures. The university Institutional
Review Board (IRB) approved all procedures, and all participants
signed informed consent forms before participation.
Data for this study were available for 39 expectant fathers who
provided prenatal demographic and neuroimaging data. Of these 39
fathers, 34 provided prenatal blood samples for oxy tocin assay, and
we excluded one oxytocin sample due to laboratory processing is-
sues. Of the 39 fathers with demographic and neuroimaging data,
37 fathers provided data on self- reported parental attunement at
3 months, and two fathers did not provide 3- month questionnaire
data. Fathers were, on average, 31.56 years old (SD = 4.25 years).
The sample was ethnically and racially diverse, with fathers report-
ing European American (33.3%), Latino (30.8%), Black or African
American (5.1%), Asian- American (25.7%), and multiracial or other
heritage (5.1%). The majority of expec tant fathers (76.9%) reported
a college degree or higher.
2.2  |  Data availability statement
The data that support the findings of this study are available from
the corresponding author upon reasonable request.
2.3  |  Procedure
During the prenatal in- laboratory visits, fathers and mothers
participated in three semi- structured discussion tasks: (a) 10-
min conversation about hopes, fears, and expectations about
birth and their transition to parenthood, (b) 10- min conversation
about projected division of childcare, and (c) 15- min conversation
about areas of conflict within the relationship. In addition to the
aforementioned discussion tasks, couples completed psychoso-
cial questionnaires, including the measures described below. At
the end of the laborator y visit, a licensed phlebotomist conducted
blood draws. Fathers returned to completed their scan session
within several weeks of their in- laboratory visit; within this sam-
ple, the lag time between the prenatal laborator y visit and the
scan range d from 1 day to 5.71 weeks , with th e majority of fathers
(32 out of 39 expectant fathers; 82.05%) completing MRI scan vis-
its within 2 weeks of the in- laboratory visit. During the MRI visit,
fathers completed the Why- How Task (Spunt & Adolphs, 2014) as
part of a larger MRI data collection protocol.
2.3.1  |  Oxytocin collection
We collected blood for the plasma oxytocin assay into a sterile
EDTA vacutainer tube. We added twenty microliters of a protease
inhibitor (Amastatin; 10 µM final concentration) to the tube to
limit oxy tocin degradation. Then, we centrifuged blood samples
for 10 min to separate plasma. We stored aliquots at −80°C and
shipped on dry ice to the University of Miami School of Medicine
Diabetes Research Institute (Armando Mendez, PI) for oxytocin
2.3.2  |  Oxytocin processing
We used enzyme- linked immunosorbent assay (ELISA) kit s (Arbor
Assays; Ann Arbor, MI) to calculate oxytocin immunoreactivity.
In our sample, the lower limit of detection was 0.8 pg/ml. In line
with prior work (Szeto et al., 2011), we extracted 2.2 ml of plasma.
Furthermore, we reconstituted in 220 µl assay buffer leading to a
10- fold concentration compared to the starting plasma volume. We
performed assays in line with manufacturer instructions. We as-
sessed samples in duplicate. The inter- assay coefficient of variation
was less than 10%. Of the MRI sample, nine fathers with ox ytocin
(27% of the sample) had oxytocin levels below detection levels fol-
lowing extraction, similar to prior research analyzing extracted ox-
ytocin (e.g., Szeto et al., 2011; Tabak et al., 2011). Consistent with
prior work, we assigned a value of 0.4 pg/ml to samples below the
detection level (Saxbe et al., 2019).
We analyzed the extrac ted oxytocin data for outliers and
normality using SPSS St atistics 26 (IBM) (M = 1.99; SD = 1.90
Skewness = 1.87). Furthermore, we truncated the data by drop-
ping outliers greater than three standard deviations from the
sample (3%). We found that extracted oxytocin data remained
positively skewed after excluding outliers (M = 1.50; SD = 1.05;
Skewness = 0.74). As a consequence, we natural log- transformed
the data to satisfy normalit y assumptions (M = 0.23; SD = 0.88;
Skewness = −0.03).
2.3.3  |  Why- how task
The Why- How Localizer Task is a standardized task for investi-
gating social cognition (Spunt & Adolphs, 2014). The task uses
Psychophysics Toolbox Version 3 (PTB- 3) and MATLAB (The
MathWorks, Inc.). The task presents individuals with 16 blocks of
items, with each block consisting of a question prompt and eight
photographs. The task consists of 42 pictures of common hand ac-
tions and 42 pictures of familiar facial expressions. The question
prompts present as pre- tested yes/no questions, with participants
given a limited amount of time for response. The task requires ap-
proximately 6 min and 30 s to complete. Participants complete the
task in a suite for a magnetic resonance imaging (MRI) scanner.
Within the scanner, participants watch stimuli while lying down
in the scanner and can use a button box programmed to allow for
yes/no responses.
The Why- How Task features a 2 (condition: Why, How) × 2 (be-
havioral c ategory: face, hand) factorial design (see Figure 1). For
the behavioral factor, participant s watch a photograph with either
a face or a hand. For the stimulus factor, participants receive a
stimulus prompt that asks either a Why question (i.e., the question
elicits theory of mind) or a How question (i.e., the question evokes
action- perception). In this stu dy, similar to Spu nt and Adolphs (2014),
we collapsed the behavioral category conditions (i.e., face, hand) and
focused on Why and How conditions. We contrasted Why > How
to specifically examine the neural correlates of theory of mind pro-
cessing. We also scored the Why- How task results behaviorally for
accuracy and response time and entered this information in SPSS for
further analysi.
2.3.4  |  fMRI data acquisition
We acquired whole- brain images on a Siemens 3 Tesla MAGNETON
Prisma System scanner. We acquired high- resolution, T1- weighted
images using a 3D Magnetization Prepared Rapid Acquisition
Gradient Echo (repetition time, 2530 ms; echo time, 3.13 ms; flip
angle, 10°), with an isotropic voxel resolution of 1 mm3. We col-
lected functional data using a T2* weighted echo- planar imaging
(EPI) with an interleaved sequence (194 2.5 mm transversal slices;
repetition time, 2000 ms; echo time, 25 ms; field of view, 192 mm2;
3.0 × 3.0 × 2.5 mm voxels, flip angle, 90 degrees). Functional scans
employed Siemens' 3D PACE prospective motion correction (and
were then pre- processed for motion further offline, described
below; Thesen et al., 2000).
We presented stimulus presentation and response recording
using MATLAB (version 2012a; MathWorks Inc., Natick, MA, USA).
We used an LCD projector to show stimuli on a rear- projection
screen. Participants made responses using their right index and mid-
dle fingers on a button box.
2.3.5  |  fMRI data analysis
We analyzed neural responses to the Why > How contrast using
FEAT (FMRI Expert Analysis Tool) Version 6.00, part of FSL
(FMRIB's Software Library, We used
FLIRT (Jenkinson et al., 2002; Jenkinson & Smith, 2001) to regis-
ter high- resolution structural and/or standard space images. For
pre- processing, we conducted motion correction with MCFLIRT
(Jenkinson et al., 20 02), non- brain removal using BET (Smith, 2002),
spatial smoothing using a Gaussian kernel of FWHM 5 mm, and
high- pass temporal filtering with Gaussian- weighted least- squares
straight- line fitting and a sigma = 50.0 s).
Next, we analyzed data using a General Linear Model (GLM)
with a multi- level mixed- ef fects design. In the GLM, the obser ved
hemodynamic response function (HRF) signal is predicted by one or
more explanatory variables that reflect a specific factor, also called
regressor or regressor of interest. The GLM is a method of model-
ing an observed neural signal in relation to one or more independent
or explanatory variables, which are also referred to as regressors
(Jenkinson & Chappell, 2018). As such, we entered each component
of the task as a regressor (Why- Face trials, Why- Hand trials, How-
Face trials, Why- Hand trials) and modeled by convolving the task
design with a double- gamma hemodynamic response function. We
defined the task periods from the first photograph in each block to
the stimulus offset of the last photograph in each block. We also
included in the model the temporal derivatives of the task regres-
sors and six motion parameters as nuisance regressors. We created
contrasts as linear combinations of the explanatory variables to look
at overall task effect s and condition effect s (e.g., Why > How). At
the individual subject level, we generated statistical maps. We then
computed brain- activity maps for each of the task effec ts and con-
dition effects combining across all subjects using a higher- level
mixed- ef fects design using FLAME to produce contrast- level activity
maps. This study focused on the Why > How contrast. All higher-
level analyses included fathers' education level and weeks pregnant
as covariates. To account for multiple comparisons, we applied clus-
ter thresholding using Gaussian Random Field theory with a cluster
size probability threshold of p < 0.05. For the main effect (Why >
How contrast), we used a cluster- forming z threshold of 3.1, a more
stringent threshold. Given that we had a smaller sample for regressor
analyses (prenatal ox ytocin, postpartum attunement), we used a less
stringent cluster- forming z threshold of 2.3 to account for low power
and avoid type II error. All coordinates repor ted are in the Montreal
Neurological Institute (MNI) standard brain space.
2.4  |  Measures
2.4.1  |  Demographics
At the prenatal visit, fathers self- reported their educational attain-
ment with the following options: high school/GED, some college,
associate degree, bachelor's degree, master's degree, professional/
doctorate). Fathers also reported on age, and mothers reported
their expected due date, from which we calculated their pregnancy
stage (number of days pregnant) at the time of their prenatal study
FIGURE 1 Conditions of the Why- How Task. The above diagram
displays the 2 × 2. factorial design of the Why- How Task with type
of question (condition: Why, How) and stimulus (condition: face,
2.4.2  |  Parental attunement
Fathers' ratings of thoughts and behaviors completed a series
of online questionnaires, including the Baby Care Questionnaire
(BCQ; Winstanley & Gattis, 2013). The BCQ is a 30- item question-
naire that asks participants how strongly they agree or disagree
with a statement related to principles that guide feeding, holding,
and soothing behaviors. Participants rate each item on a Likert
Scale from 1 (strongly disagree) to 4 (strongly agree). The BCQ
generates two subscales, “Structure” and “Attunement,” which
reflect parents' endorsement of more structured (reliant on rou-
tines and schedules) versus more attuned (responsive to in- the-
moment infant cues) parenting style principles. We specifically
assessed the Attunement scale for this project. BCQ Attunement
subscale scores are generated by averaging scores on attunement
items, such as “Babies benefit from physical contact with parent s
when they wake during the night,” Parent(s) should find a pattern
of feeding/eating that suits their baby,” and “Responding quickly to
a crying baby leads to less crying in the long run.” A prior validation
study showed that the BCQ attunement subscale had acceptable
internal consistency (Cronbach's alpha >0.70), acceptable test–
retest reliability (>0.70), and external validity (i.e., endorsement
of attunement principle was correlated with higher bed- sharing,
breastfeeding, and holding; Winstanley & Gattis, 2013). We com-
puted internal consistency on our sample and found acceptable
reliability in our sample's BCQ Attunement scores (Cronbach's
alpha = 0.70).
We present descriptive statistics in Table 1. Zero- order correlations
between study variables are shown in Table 2. Fathers' education
and age were positively correlated (r (39) = 0.57, p < 0.01). Age cor-
related with accuracy during How trials (r (39) = 0.35, p < 0.05).
Fathers' accuracy during Why trials correlated with fathers' accu-
racy during How trials (r (38) = 0.78, p < 0.01). Fathers' response
time during Why trials correlated with fathers' response time during
How trials (r (38) = 0.9 3, p < 0.01). We also found a trending negative
correlation between extracted oxytocin and response time during
How trials (r (32) = −0.35, p = 0.05). Of interest, we found a trending
significance between prenatal ox ytocin and postpartum attunement
(r (31) = 0.33, p = 0.069, see Figure 2).
3.1  |  Behavioral results
Within this sample, we found behavioral results consistent with
Spunt and Adolphs (2014). The percent accuracy on How (M = 94%,
SD = 6.2%) trials differed from Why (M = 91%, SD = 7.3%) trials,
t (38) = 4.09, p < 0.01, such that participants responded less ac-
curately to Why trials. Response time to How trials (M = 818 ms.,
SD = 134 ms.) differed from Why trials (M = 900 ms., SD = 155), t
(37) = −8.38, p < 0.01, such that par ticipants responded more slowly
during Why trials.
3.2  |  fMRI results
3.2.1  |  Covariates: Associations between
education and weeks pregnant during theory of mind
Higher educational attainment predicted greater signal change in
theory of mind regions during the Why > How contrast. When we
added education level as a regressor into the Why > How contrast
(see Table 3 and Figure 3), signal changes associated with this co-
variate were found in four clusters, including the right precuneus,
bilateral precentral g yri, left postcentral gyrus, and right inferior
temporal gyrus. Given these findings, we included education as a
covariate in all analyses with cross- subject regressors. The preg-
nancy stage was also included as a covariate in all analyses to control
for potential differences resulting from men's partners' pregnancy
stage. When we added weeks pregnant as a regressor into the Why
> How contrast (see Table 4 and Figure 4), activation occurred in a
cluster near the cingulate, largely comprised of white matter. Thus,
we mean- centered education level and pregnancy stage (i.e., weeks
pregnant) and included as confound regressors in all analyses.
TABLE 1 Descriptive statistics of sample demographics and self-
reported measures.
Variable Subcategory Value/mean (%/SD)
Education High school/GED 1 (2.6%)
Some college 7 (17.9%)
Associate degree 1 (2.6%)
Bachelor's degree 12 (30.8%)
Master's degree 11 (28.2%)
Professional/doctorate 7 (17.9%)
Race European American 13 (33.3%)
Latino 12 (30.8%)
Black or African
2 (5.1%)
Asian- American 10 (25.7%)
Multiracial or other 2 (5.1%)
Age at prenatal
31.56 (4.25), 23– 41
Extr acted
0.23 (0. 88),
−0.92– 1.92
Weeks pregnant 27.84 (4.71),
2 0 . 3 7 – 3 8 . 5 2
Weeks between
and MRI
1.63 (1.15),
0 . 1 4 – 5 . 7 1
2.91 (0.31),
2 . 4 2 – 3 . 7 7
Weeks pregnant is the self- reported time fathers reported their female
partners were pregnant on the day of the laboratory visit.
3.2.2  |  Hypothesis 1: Neural activation during main
effect of task
We partially replicated the key neural findings from the Spunt and
Adolphs (2014) paper. The full list of brain regions activated by
Why > How contrast is listed in Table 5. A visual representation
of the brain regions activated by Why > How contrast is shown in
Figure 5. When contrasting activation during Why trials, compared
with How trials (i.e., Why > How contrast), we found signal change
in regions from the original group- level results of the study: left
dorsomedial prefrontal cortex, left orbitofrontal cortex, right tem-
poroparietal junction (angular gyrus), bilateral posterior cingulate
cortex, and right anterior superior temporal sulcus. These results
largely replicate findings from the original Why- How paper (Spunt
& Adolphs, 2014). However, during the Why > How contrast, this
study found additional signal change in several areas not found in
the original task study: left amygdala, left dorsolateral prefrontal
cortex, left caudate, left thalamus, and right posterior superior tem-
poral gyrus.
3.2.3  |  Hypothesis 2: Associations between
oxytocin and BOLD signal during theory of mind
Prenatal oxytocin was added as a between- subjects regressor in the
group- level analysis of the Why > How contrasts. Higher prenatal
TABLE 2 Bivariate correlation of main study variables.
Variable 1 2 345678
1. Age
2. Education 0.57***
3. Weeks pregnant 0.22 0.29*
4. Prenatal oxytocin −0.05 −0.10 0.23
5. Parental attunement −0.10 0.15 −0.10 0.33*
6. Why trials accuracy 0.15 0.13 0.02 −0 .11 −0.18
7. How trials accuracy 0.35** 0.19 0.14 −0.09 0.21 0.78***
8. Why trials RT 0.28*0.1 2 −0.03 −0.28 0.11 0.25 0.05
9. How trials RT 0.17 0.07 −0.08 −0.35** −0.16 0.28*0.01 0.93***
Weeks pregnant is the self- reported time fathers reported their female partner s were pregnant including the time between the laboratory visit and
the MRI visit.
*p < 0.10; **p < 0.05; ***p < 0.01.
FIGURE 2 Scatterplot of prenatal plasma oxytocin and postpartum self- reported parental attunement.
Prenatal plasma oxytocin
Postpartum self-reported parental attunement
Region L/R k
MNI coordinates
x y z Z
Precuneus R2127 18 −62 42 4. 55
R - 30 −40 46 4.38
R - 10 −66 60 4.05
R - 54 −52 50 3 .93
R - 32 −5 0 56 3.87
R - 34 −56 64 3.85
Precentral gyrus R824 10 −32 50 3.76
L - −22 −60 56 3.73
R - 2 −42 54 3.71
R - 4 −38 54 3.69
R - 8 −34 44 3. 52
L - −16 −58 52 3 .17
Postcentral gyrus L632 −46 −28 52 4.02
L - −62 −38 44 3.91
L - −62 −42 44 3.8
L - −48 −26 44 3 .52
L - −32 −26 42 3.45
L - −40 −18 60 3 .23
Inferior temporal gyrus R494 50 −58 −8 3.84
R - 38 −56 23.46
R - 30 −6 4 03.36
R - 18 −60 14 3.34
R - 24 −7 4 23.24
R - 28 −58 −2 3.24
All peak s survived a whole- brain search thresholded at a voxel- wise family- wise error rate of 0.05
and a cluster corrected at z = 2.3. x, y, z = Montreal Neurological Institute (MNI) coordinates in the
left- right anterior- posterior, and inferior- superior dimensions, respectively (n = 39).
TABLE 3 Neural regions positively
correlated with education during the Why
> How contrast.
FIGURE 3 Group- level results of the Why > How contrast with education as a regressor. Sagittal, coronal, and axial view of whole- brain
activation during the Why > How contrast. Analyses cluster corrected at z = 2.3 (n = 39).
plasma oxytocin predicted greater signal in theory of mind regions
during the Why > How contrasts. Specifically, signal change emerged
in eight clusters, including bilateral temporoparietal junction (angular
gyrus), bilateral supramarginal gyrus, bilateral dorsolateral prefrontal
cortex, bilateral paracingulate gyrus, bilateral intracalcarine cortex,
bilateral posterior cingulate gyrus, and right superior temporal gyrus
(see Table 6 and Figure 6).
3.2.4  |  Hypothesis 3: Associations between
postpartum attunement and activation
during theory of mind
We also explored whether signal change in theory of mind regions
during the Why > How contrast predicted greater endorsement of
an attuned parenting philosophy at 3 months postpartum. We found
that postpartum attunement was associated with signal change for
the Why- How contrast in two regions including por tions of the
bilateral dorsolateral prefrontal cortex and right posterior parahip-
pocampal gyrus (see Table 7 and Figure 7).
3.2.5  |  Exploratory Analyses: Overlap between
oxytocin and postpartum attunement activation
during theory of mind
To follow- up on the above- repor ted findings, we assessed whether
there were areas of significant overlap in signal change asso ciated with
both prenatal oxytocin and postpartum attunement during the Why
> How contrast. Specifically, we used fslmaths (Smith et al. 2004) to
mask the thresh_zstat image of the Why > How contrast with oxy-
tocin as a regressor (i.e., hypothesis 2) with the thresh_zstat image of
the Why > How contrast with postpartum attunement as a regressor
(i. e., hypothesis 3). The re sul ts of this ma ski ng (Figure 8) reveal ed over-
lapping activation in a mostly right- lateralized region in the prefrontal
cortex, mostly composed of the dorsolateral prefrontal cortex.
Region L/R k
MNI coordinates
x y z Z
White matter/cingulate gyrus R400 26 −46 12 3.34
R - 36 −60 43.32
R - 32 −62 83.31
R - 16 −38 28 3.22
R - 20 −48 18 3.21
R - 34 −68 −2 3.16
All peak s survived a whole- brain search thresholded at a voxel- wise family- wise error rate of 0.05
and a cluster corrected at z = 2.3. x, y, z = Montreal Neurological Institute (MNI) coordinates in the
left- right anterior- posterior, and inferior- superior dimensions, respectively (n = 39).
TABLE 4 Neural regions positively
correlated with weeks pregnant during
Why > How contrast.
FIGURE 4 Group- level results of the Why > How contrast with weeks pregnant as a regressor. Sagittal, coronal, and axial view of
whole- brain activation during the Why > How contrast. Analyses cluster corrected at z = 2.3 (n = 39).
Region L/R k
MNI coordinates
x y z Z
Orbitofrontal cor tex L6372 −48 16 −22 6.45
L - −58 −8 −14 6.4
L - −42 28 −14 6.37
L - −56 4−16 6. 29
L - −52 24 12 6.22
L - −48 30 −6 6.21
Dorsomedial prefrontal cortex L5759 −6 56 38 7. 2 2
L - −8 64 26 7.2
L - −4 44 50 7.1 8
L - −10 60 32 7.18
L - −2 48 −16 7.16
L - −10 44 46 7
Posterior cingulate cor tex R1836 0−50 30 6.27
L - −4 −60 34 6.08
L - −2 −60 30 5.93
L - −4 −50 22 5.64
L - −10 −52 44.31
R - 14 −48 63.64
Anterior superior temporal sulcus R684 58 2−22 5.31
R - 60 −6 −18 5.25
R - 62 −2 −22 5.03
R - 54 8−22 4.98
R - 50 −8 −12 4.89
R - 48 −14 −10 4.8
Amygdala L261 −26 −18 −18 5.01
L - −20 −1 0 −14 4.96
L - −28 −6 −2 2 4.19
L - −36 −16 −2 2 3.43
Dorsolateral prefrontal cortex L229 −42 14 54 4.22
L - −36 24 52 4.22
L - −34 18 50 4.17
Temporoparietal junction (angular gyrus) R161 48 −58 24 4. 24
R - 52 −62 28 4.02
R - 58 −56 26 3.64
Caudate L159 −1 2 14 12 4.65
L - −10 −4 12 3.29
L - −14 14 03.25
Thalamus L102 −4 −1 2 64.31
L - −2 −16 12 3.91
Posterior superior temporal gyrus R100 64 −34 43.94
R - 54 −30 23.86
R - 46 −34 −2 3.78
All peak s survived a whole- brain search thresholded at a voxel- wise family- wise error rate of 0.05
and a z- threshold of 3.1. x, y, z = Montreal Neurological Institute (MNI) coordinates in the left- right
anterior- posterior, and inferior- superior dimensions, respectively (n = 39).
TABLE 5 Group- level results of the
Why > How contrast.
In this study, we partially replicated the results from the Spunt
and Adolphs (2014) paper within a sample of expectant fathers.
Consistent with our hypotheses, we found that theor y of mind pro-
cessing during the Why > How contrast was associated with signal
change in mentalizing net work structures, including both anterior
and posterior cortical midline structures and the right temporopa-
rietal junction (angular gyrus). We also found activation to Why >
How in a few additional regions, including the left dorsolateral pre-
frontal cortex, right posterior superior temporal gyrus and subcor-
tical regions (left caudate, left thalamus, lef t amygdala). Moreover,
we found that prenatal oxytocin in expectant- fathers during the
Why > How contrast was associated with similar signal change as
the Why > How contrast, including frontal and posterior cortical
midline structures as well as the bilateral temporoparietal junction
(angular gyrus). Signal change in the bilateral dorsolateral prefrontal
cortex and right posterior hippocampal gyrus during the Why > How
contrast predicted greater self- reported beliefs and cognitions re-
garding at tuned parenting styles approximately 6 months later when
infants were 3 months of age.
Although not originally found by Spunt and Adolphs (2014), we
observed activation in the left amygdala and left caudate during
the Why > How contrast in expectant fathers. Interestingly, social
neuroscientists have posited that regions associated with process-
ing emotional salience (e.g., fearful faces) may foster higher order
empathic behavior by making individuals aware of the suffering of
others (Marsh, 2016). Within parenting, the amygdala may promote
mothers' ability to detect biologically salient information in the care-
giving environment (Swain et al., 2014). The caudate is a reward re-
gion and is theorized to support mothers' motivation to engage with
their infants (Noriuchi et al., 2008). Men's testosterone across the
transition to parenthood, which may reflect parenting investment,
has been associated with increased caudate activation (Kuo et al.,
2012). Although preliminary, our findings suggest that expectant
fathers are engaging both cortical and subcortical regions when
completing the Why- How theory of mind task; further research can
investigate whether these patterns of activation are dynamic in new
Although not the focus of the study, fathers' educational attain-
ment was associated with Why > How activation in areas involved
in vision and the processing of words (e.g., inferior temporal gyrus;
Dien et al., 2013), the mirror neuron system (e.g., precentral gyrus,
postcentral gyrus; Wu et al., 2017), and theory of mind (e.g., pre-
cuneus; Saxe et al., 2006). In the parenting network, the mirror
neuron system and the theory of mind network work together to
support a parents' ability to quickly resonate with infant behaviors
(i.e., mirror neuron system), consider intentions (i.e., theory of mind
network), and provide caregiving behavior that syncs with the child's
needs (Feldman, 2015). Extensive research suggest s that fathers
with higher education levels use more advanced vocabulary, which
is linked with children's emotion regulation language development
(Cabrera et al., 2007; Pancsofar & Vernon- Feagans, 2010). Thus, our
findings might reflect that more educated fathers may exhibit more
comfort or engagement with the Why- How Task, which requires
participants to read questions quickly.
Additionally, fathers' prenatal oxy tocin levels predicted acti-
vation in areas of the brain that support embodied simulation and
theory of mind. Of note, we collected prenatal oxy tocin following a
laboratory visit that included partner interaction and questions rel-
evant to the transition to parenthood. Ac tivation during the Why >
How contrast was associated with prenatal oxytocin in a region that
included the bilateral temporoparietal junction (angular gyrus) and
bilateral supramarginal gyrus, which compose the inferior parietal
FIGURE 5 Group- level results of the Why > How contrast with education and weeks pregnant as covariates. Sagittal, coronal, and axial
view of whole- brain activation during the Why > How contrast. Analyses cluster corrected at z = 3.1 (n = 39).
lobule (IPL). The angular gyrus, which is part of the temporopari-
etal junction, has been implicated in theory of mind and mentalizing
tasks (for review, see Seghier, 2013). Concerning parenting, the IPL
is part of the embodied simulation network as well as the mirror neu-
ron system (Feldman, 2015). The mirror neuron system supports an
individual's ability to observe and mimic the emotions and behaviors
of others (for review, see Rizzolatti & Craighero, 2004). The embod-
ied simulation network supports a parents' ability to mentally sim-
ulate the behavior of the infants and consider their intentions and
needs (Feldman, 2015). The Why > How contrast was also related
to oxytocin in areas that support social cognition, including the pos-
terior cingulate cortex, a key region of the Default Mode Network
(Li et al., 2014). Given the minimal research on oxy tocin in expect-
ant fathers and the preliminary evidence we found of an association
between oxytocin and social cognition, we recommend that future
studies investigate oxytocin and measures of social cognition across
the transition to fatherhood.
Prenatal activation to the Why > How contrast in regions that
support social cognition, behavioral organization, and emotion
regulation predicted fathers' postpartum endorsement of a more
attuned parenting style, as assessed 3 months following the bir th.
Specifically, attunement was associated with activation in two clus-
ters, including the bilateral dorsolateral prefrontal cortex and right
posterior parahippocampal gyrus. The dorsolateral prefrontal cor-
tex has been activated in studies assessing attention (Doricchi et al.,
2010; Shulman et al., 2010), executive control (Wagner et al., 2001),
and theory of mind (Kobayashi et al., 2007). The parahippocampal
gyrus is part of the paralimbic region and underlies context and
memory processing (Bohbot et al., 2015). With regard to parenting,
the dorsolateral prefrontal cortex is posited to be part of the pa-
rental emotion- regulation/executive network (Feldman, 2015). The
emotion- regulation/executive network is theorized to contribute to
parents' management of emotions, attention, multi- tasking, and or-
ganization (Feldman, 2015; Swain et al., 2014). Furthermore, in a lon-
gitudinal study of cortical plasticity in new fathers, Kim et al. (2014)
found that fathers' show an increase in gray matter volume in the lat-
eral prefrontal cortex across the first and four th month postpartum.
TABLE 6 Neural regions positively correlated with prenatal
oxytocin during Why > How contrast.
Region L/R k
MNI coordinates
x y z Z
TPJ/SM G L3882 −48 −56 46 4.68
L - −58 −48 40 4.68
L - −50 −5 0 38 4.66
L - −42 −52 54 4.54
L - −52 −38 42 4.06
L - −2 0 −74 42 4.05
TPJ/SM G R1481 44 −56 50 4.17
R - 46 −50 32 3.67
R - 34 −80 32 3.94
R - 40 −60 46 3.93
R - 48 −44 56 3.87
R - 52 −46 40 3.63
R817 34 34 30 3.57
R - 44 30 26 3.45
R - 44 32 38 3.2
R - 52 36 26 3.2
R - 36 18 36 3 .16
R - 22 50 24 3.25
L811 −36 −2 54 3.76
L - −38 20 50 3.60
L - −50 26 32 3 .47
L - −36 26 44 3 .47
L - −38 32 28 2 .97
L - −50 32 24 2 .96
R707 12 30 26 3.58
L - −10 42 30 3.29
R - 2 14 56 3.05
L - −6 28 44 3.14
R - 8 60 28 3.22
- 0 30 22 3.09
R561 2−64 16 3.27
L - −16 −62 03.23
L - 12 −7 8 63.16
L - −1 2 −72 14 3.16
L - −8 −68 10 3.04
L - −1 8 −72 23.03
R429 2−36 46 3.71
R - 2 −44 50 3.57
L - −2 −30 44 3.29
L - −10 −24 42 3.25
L - −1 2 −20 40 3.22
L - −4 −2 2 46 2.92
Region L/R k
MNI coordinates
x y z Z
R340 48 −20 −4 4.04
R - 62 −2 0 23.21
R - 62 −2 8 23.10
R - 70 −2 6 −4 3.88
R - 60 −1 2 −10 3.44
R - 60 −2 2 −18 3.28
All peak s survived a whole- brain search thresholded at a voxel- wise
family- wise error rate of 0.05 and a z- threshold of 2.3. x, y, z = Montreal
Neurological Institute (MNI) coordinates in the left- right anterior-
posterior, and inferior- superior dimensions, respectively (n = 33). TPJ/
SMG =Temporoparietal Junction (Angular Gyrus)/Supramarginal Gyrus.
TABLE 6 (Continued)
Prior research suggests that the paralimbic region, which includes
the parahippocampal gyrus, is involved in processing own- infant
stimuli (Swain, 2008). We found that fathers' neural activation in
response to a theory of mind task during their partner's pregnancy
predicted their subsequent principles and cognitions about attuned
parenting at 3 months after birth. For example, they endorsed be-
liefs that infants benefit from eating on demand and being soothed
to sleep rather than kept on strict schedules. More attuned par-
enting may require more competent mentalizing abilities because
parents need to “read” infants' cues rather than follow pre- ordered
routines. It is intriguing that fathers' prenatal theory of mind activa-
tion predicted their subsequent beliefs about this style of parenting.
Both prenatal oxytocin and postpartum endorsement of attuned
parenting styles showed similar activation in the dorsolateral pre-
frontal cortex during the Why > How contrast. As mentioned, the
dorsolateral prefrontal cortex is implicated in executive func tioning
and is posited to support parent attention and emotion regulation
(Feldman, 2015). The dorsolateral prefrontal cortex has not been
previously associated with oxytocin within the context of parenting.
However, research assessing the neural underpinnings of social sup-
port found that oxytocin enhanced the benefits of romantic partner
touch during shock, and oxytocin was associated with increased ac-
tivity in the middle frontal gyrus, which is par t of the dorsolateral
prefrontal cortex (Kreuder et al., 2018). Relatedly, Weisman et al.
FIGURE 6 Group- level results of the Why > How contrast with oxytocin as the main regressor and education and weeks pregnant as
covariates. Sagittal, coronal, and axial view of whole- brain activation during the Why > How contrast. Analyses cluster corrected at z = 2.3
(n = 33).
Region L/R k
MNI coordinates
x y z Z
Dorsolateral prefrontal cortex R2120 32 34 28 3.98
L - −1 2 44 36 3.74
R - 30 840 3.62
R - 36 22 38 3.58
L - −8 56 24 3.55
R - 14 52 26 3.5
Posterior parahippocampal gyrus R498 14 −34 −4 3.68
R - 10 −36 −4 3.66
R - 2 −42 03 .57
R - 18 −3 4 −14 3.15
R - 0 −56 16 3.09
R - 16 −3 0 23.06
All peak s survived a whole- brain search thresholded at a voxel- wise family- wise error rate of 0.05
and a z- threshold of 2.3. x, y, z = Montreal Neurological Institute (MNI) coordinates in the left- right
anterior- posterior, and inferior- superior dimensions, respectively (n = 37).
TABLE 7 Neural regions positively
correlated with postpartum attunement
during Why > How contrast.
(2014) found a link between oxy tocin administration and changes in
testosterone levels linked with enhanced fathering behaviors (e.g.,
vocal synchrony, positive affect, greater touch). Within this sample,
we have found that paternal postpartum oxytocin levels are asso-
ciated with more father– infant touch during parenting interaction
(Morris et al., under review). This study found associations in the
dorsolateral prefrontal cortex during theor y of mind with prenatal
oxytocin and fathers' greater endorsement of attunement as a post-
partum parenting principle. Given that all variables were measured
once, we cannot infer directionality or causalit y between prenatal
oxytocin and theor y of mind processing and postpar tum attunement
principles. As a natural extension of our work, future studies can
assess trajectories of oxytocin, theory of mind processing, and par-
enting across the transition to parenthood to assess longitudinal
This study has a number of limitations. O ur sample was ethnically
diverse, but participants were largely from highly educated back-
grounds. Although larger than other samples in the neurobiology
of parenting literature, the sample size was small (39 fathers with
MRI data; 33 with oxytocin data; 37 with postpartum attunement
data). An additional limitation is that for several of our fathers, oxy-
tocin levels were either missing (due to refusal or sample processing
FIGURE 7 Group- level results of the Why > How contrast with the Baby Care Questionnaire (BCQ) attunement scale as the main
regressor and with education and weeks pregnant as covariates. Sagittal, coronal, and axial view of whole brain activation during the Why >
How contrast. Analyses cluster corrected at z = 2.3 (n = 37).
FIGURE 8 Overlapping activity during the Why > How contrast with oxytocin and postpartum attunement as the main regressors.
Sagitt al, coronal, and axial view of whole- brain activation. Analyses cluster corrected at z = 2.3 and a p value of .05 (n = 33 for ox ytocin,
n = 37 for BCQ attunement).
issues) or undetectable (due to overly low levels). Furthermore,
we note that this study only involved one neuroimaging timepoint
during the transition to parenthood. The extant literature supports
that neural regions involved in social cognition can develop and
change over time (Blakemore, 2008; Dumontheil et al., 2010). Future
research that incorporates multiple scans throughout the transition
to parenthood can better characterize the fathering brain. As the
first study to examine the theor y of mind processing, oxytocin, and
postpartum parenting attitudes together in one investigation, this
work would benefit from replication and extension prior to further
Despite the aforementioned limitations, this study makes sev-
eral important contributions to research on parenting and child
development. This study extends prior work on the neurobiology
of fatherhood by using a standardized task to measure the theory
of mind in expectant fathers. This is also the first study examining
prenatal oxytocin levels in tandem with brain ac tivation in expect-
ant fathers. Our sample of expectant fathers was racially and eth-
nically diverse and contributed longitudinal, multimodal data. Our
results suggest that oxytocin may support social cognition process-
ing during the prenatal period. Additionally, fathers who showed
greater prenatal neural theory of mind activation subsequently
endorsed more attuned parenting styles at 3 months postpartum.
These findings further the current understanding of neurobiolog-
ical factors that support fathers' preparation for parenthood and
highlight future directions for the parenting brain research.
This work wa s supported by N ational Scien ce Foundation (NS F) Career
Award (1552452; Saxbe), National Science Foundation Grant/Award
(DGE- 1842487; Cárdenas), and a Health Policy Research Scholars
Award funded by the Robert Wood Johnson Foundation (Cárdenas).
Portions of this work were presented at the International Society of
Developmental Psychobiology (ISDP) Conference; we thank ISDP and
their sponsors (i.e., NICHD; Columbia University's Nurture Science
Project) for granting a Travel Award to Sofia Cárdenas.
The authors have no conflic t of interests to report.
Sofia I. Cardenas Conceptualization (equal); Formal Analysis
(equal); Investigation(supporting); WritingOriginal Draft
Preparation (lead); Sarah A. Stoycos: Investigation (equal)); Project
Administration (leading); Writing— Review & Editing (supporting);
Pia Sellery: Investigation (supporting); Project Administration
(supporting); WritingReview & Editing (supporting); Narcis
Marshall: Investigation (supporting); Writing— Review & Editing
(supporting); Hannah Khoddam: Investigation (supporting);
Project Administration (supporting); Writing— Review & Editing
(supporting); Jonas Kaplan: Formal Analysis (equal); Writing—
Review & Editing (supporting); Diane Goldenberg: Investigation
(equal); Project Administration (supporting); Darby E. Saxbe:
Conceptualization (equal); Funding acquisition (lead); Writing—
Review & Editing (supporting).
Sofia I. Cardenas
Darby E. Saxbe
Abraham, E., Hendler, T., Shapira- Lichter, I., Kanat- Maymon, Y., Zagoor y-
Sharon, O., & Feldman, R . (2014). Father's brain is sensitive to child-
care experiences. Proceedings of the National Academy of Sciences
of the United States of America, 111(27), 9792– 9797. https://doi.
org /10.1073/pnas.14025 69111
Atzil, S., Hendler, T., Zagoor y- Sharon, O., Winetraub, Y., & Feldman, R.
(2012). Synchrony and specificity in the maternal and the paternal
brain: Re lations to oxyt ocin and vasopre ssin. Journal of th e American
Academy of Child and Adolescent Psychiatry, 51(8), 798– 811. https://
Bart z, J. A., Zaki, J., Bolger, N., & Ochsner, K. N. (2011). Social ef-
fects of oxytocin in humans: Context and person matter. Trend s
in Cognitive Sciences, 15(7):301– 309.
Blakemore, S. J. (2008). Development of the social brain during adoles-
cence. Quarterly Journal of Experimental Psychology, 61(1), 40– 49.
htt ps://doi .org /10.1080/17470 21070 15 08715
Bohbot, V. D., Allen, J. J. B., Dagher, A., Dumoulin, S. O., Evans, A. C.,
Petride s, M., Kalina, M., Stepankova, K., & Nadel, L. (2015). Role of
the parahippocampal cortex in memory for the configuration but
not the identity of objects: Converging evidence from patients with
selec tive thermal le sions and fMRI. Fron tiers in Human Neu roscience,
9, 431. 2015.00431
Cabrera, N. J., Shannon, J. D., & Tamis- LeMonda, C. (20 07). Fathers' in-
fluence on their Children's cognitive and emotional development:
From toddlers to Pre- K. Applied Developmental Science, 11(4), 208–
213. .1080/10888 69070 1762100
Cabrera, N. J., Volling, B. L., & Barr, R. (2018). Fathers are parents, too!
Widening the lens on parenting for children's development. Child
Development Perspectives, 12(3), 152– 157.
Cárdenas, E. F., Kujawa, A., & Humphreys, K. L. (2019). Neurobiologic al
changes during the peripartum period: Implications for health and
behavior. Social Cognitive and Affective Neuroscience, 15(10), 1097–
Carpendale, J. I., & Chandler, M. J. (1996). O n the dis tinction bet ween
false belief unders tanding and subscribing to an interpretive the-
ory of mind. Child Development, 67(4), 1686– 1706. https://doi.
org /10.1111/j.1467- 8624.1996.t b018 21.x
Cutting, A . L., & D unn, J. (1999). Theory of mind, emotion understand-
ing, language, and family background: Individual differences and
interrelations. Child Development, 70(4), 853– 865. https://doi.
org /10.1111/1467- 8624.0 0061
Dien, J., Brian, E. S., Molfese, D. L., & Gold, B. T. (2013). Combined ERP/
fMRI evidence for early word recognition effects in the posterior
inferior temporal g yrus. Cortex, 49(9), 2307– 2321. https://doi.
Domes, G., Heinrichs, M., Michel, A., Berger, C., & Herpertz, S. C. (2007).
Oxytocin improves “mind- reading” in humans. Biological Psychiatry,
61(6), 731– 733. ych.2006.07.015
Doricchi, F., Macci, E., Silvetti, M., & Macaluso, E. (2010). Neural cor-
relates of the spatial and expectancy components of endogenous
and stimulus- driven orienting of at tention in the Posner task .
Cerebral Cortex, 20(7), 1574– 1585.
r/b hp215
Dumontheil, I., Apperly, I. A., & Blakemore, S. J. (2010). Online usage
of theor y of mind continues to develop in late adolescence.
Developmental Science, 13(2), 331– 338. ht tps:// .1111/
j . 1 4 6 7 - 7 6 8 7 . 2 0 0 9 . 0 0 8 8 8 . x
Edelstein, R., Chopik, W. J., Saxbe, D. E., Wardecker, B. M., Moors, A.
C., & LaBelle, O. P. (2017). Prospective and dyadic associations
between expectant parents' prenatal hormone changes and post-
partum parenting outcomes. Developmental Psychobiology, 59(1),
77– 90.
Ereky- Stevens, K. (2008). Associations between mothers' sensitivity to
their infants' internal states and children's later understanding of
mind and emotion. Infant and Child Development, 17 (5), 527543.
htt ps://doi.o rg/10.10 02/icd.572
Feldman, R. (2015). The adaptive human parental brain: Implic ations for
children's social development. Trends in Neurosciences, 38(6), 387
Feldman , R., & Bakerman s- Kranenbu rg, M . J. (2017). Ox ytoci n: A parent-
ing hormone. Current Opinion in Psychology, 15, 13– 18. https://doi.
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 ox ytocin following
parent- infant contact. Psychoneuroendocrinology, 35(8), 1133– 1141. uen.2010.01.013
Frith, C. D., & Frith, U. (2006). The neural basis of mentalizing. Neuron,
50(4), 531– 534.
Frith, U., & Frith, C. D. (2003). Development and neurophysiology
of mentalizing. Philosophical Transactions of the Royal Society B:
Biological Sciences, 358(1431), 459– 473. ht tps:// /10.1098/
Gallagher, H. L., & Frith, C. D. (2003). Functional imaging of ‘theor y
of mind’. Trends in Cognitive Sciences, 7(2), 77– 83. https://doi.
o r g / 1 0 . 1 0 1 6 / S 1 3 6 4 - 6 6 1 3 ( 0 2 ) 0 0 0 2 5 - 6
Gallagher, H. L., Happé, F., Brunswick, N., Fletcher, P. C., Frith, U., &
Frith, C. D. (2000). Reading the mind in cartoons and stories:
An fMRI study of ‘theory of mind’ in verbal and nonverbal tasks.
Neuropsychologia, 38(1), 11– 21.
- 3 9 3 2 ( 9 9 ) 0 0 0 5 3 - 6
George, M. R. W., Cummings, E. M., & Davies, P. T. (2010). Positive as-
pects of fathering and mothering, and children's attachment in
kindergarten. Early Child Development and Care, 180( 1 – 2 ) , 1 0 7 1 1 9 .
htt ps://doi.o rg/10.10 80/03 004 43 090 3414752
Gettler, L. T. (2014). Applying socioendocrinology to evolutionary mod-
els: Fatherhood and physiology. Evolutionary Anthropology, 23(4),
146– 160.
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. ht tps:// /10.1016/j.biops
Gregor y, R., Cheng, H., Rupp, H. A., Sengelaub, D. R., & Heiman, J.
R. (2015). Oxytocin increases V TA activation to infant and sex-
ual stimuli in nulliparous and postpartum women. Hormones
and Behavior, 69, 8288.
Hoekzema, E., Barba- Müller, E., Pozzobon, C., Picado, M., Lucco, F.,
García- García, D., Soliva, J. C., Tobeña, A., Desco, M., Crone, E.
A., Ballesteros, A ., Carmona, S., & Vilarroya, O. (2017). Pregnancy
leads to long- lasting changes in human brain structure. Nature
Neuroscience, 20(2), 287– 296.
Jenki nso n, M., Ba nni ste r, P., Br ady, M., & Smit h, S. (20 02). Imp roved opti-
mization for the robust and accurate linear registration and motion
correction of brain images. NeuroImage, 17(2), 825– 841. https://
d o i . o r g / 1 0 . 1 0 1 6 / s 1 0 5 3 - 8 1 1 9 ( 0 2 ) 9 1 1 3 2 - 8
Jenkinson, M., & Chappell, M. (Eds.), (2018). Introduction to neuroimaging
analysis, (1st ed.) New York, NY: Oxford University Press.
Jenkinson, M., & Smith, S . (2001). A global optimization method for ro-
bust affine regist ration of brain images. Medical Image Analysis, 5(2),
1 4 3 – 1 5 6 . h t t p s : / / d o i . o r g / 1 0 . 1 0 1 6 / s 1 3 6 1 - 8 4 1 5 ( 0 1 ) 0 0 0 3 6 - 6
Keysar, B., Lin, S ., & Barr, D. J. (20 03). Limits on theory of mind use in
adults. Cognition, 89(1), 25– 41.
- 0 2 7 7 ( 0 3 ) 0 0 0 6 4 - 7
Khoddam, H., Goldenberg, D., Stoycos, S. A., Horton, K . T., Marshall,
N., Cárdenas, S. I., Kaplan, J., & Saxbe, D. (2020). How do expect-
ant fathers respond to infant cry? Examining brain and behav-
ioral responses and the moderating role of testosterone. S ocial
Cognitive and Affective Neuroscience, 15(4), 437446. https://doi.
org /10.1093/sc an/nsaa051
Ki m, P., Ri go , P., Ma yes , L. C., Fel dma n, R. , Le ck man , J. F., & Sw ai n, J. E. (2 014 ).
Neural plasticit y in fathers of human infant s. Social Neuroscience, 9(5),
522– 535. htt ps://doi .org /10.1080/17470 919.2014.933713
Kobayashi, C ., Glover, G. H., & Temple, E. (20 07). Children's and
adults' neural base s of verbal and nonverbal “theory of mind”.
Neuropsychologia, 45( 7), 1522– 1532.
neuro psych ologia.2006.11.017
Kreuder, A. K., Wassermann, L., Wollseifer, M., Ditzen, B., Eckstein, M.,
Stoffel- Wagner, B., Hennig, J., Hurlemann, R., & Scheele, D. (2018).
Oxytocin enhances the pain- relieving effect s of social support in
romantic couples. Human Brain Mapping, 40(1), 242– 251. https://
Kuo, P. X., Carp, J., Light, K. C., & Grewen, K. M. (2012). Neural re-
sponses to infants linked with behavioral interactions and testos-
terone in fathers. Biological Psychology, 91(2), 302– 3 06. https://doi.
org/10.1016/j.biops ycho.2012.08.002
Lalonde, C. E., & Chandler, M. J. (1995). False belief understanding goes
to school: On the social- emotional consequences of coming early
or late to a first theor y of mind. Cognition and Emotion, 9(2– 3), 167
185. 93950 8409007
Leake, R. D., Weitzman, R . E., Glatz, T. H., & Fisher, D. A. (1981). Plasma
oxytocin concentrations in men, nonpregnant women, and preg-
nant women before and during spontaneous labor. The Journal of
Clinical Endocrinology and Metabolism, 53(4), 730– 733. https://doi.
o r g / 1 0 . 1 2 1 0 / j c e m - 5 3 - 4 - 7 3 0
Lefevre, A., Mottolese, R., Dirheimer, M., Mottolese, C., Duhamel, J.- R.,
& Sirigu, A. (2017). A comparison of methods to measure central
and peripheral oxytocin concentrations in human and non- human
primates. Scientific Reports, 7(1), 17222.
s 4 1 5 9 8 - 0 1 7 - 1 7 6 7 4 - 7
Li, W., Mai, X., & Liu, C. (2014). The default mode network and social
understanding of others: What do brain connectivity studies tell
us? Frontier s in Human Neurosci ence, 8, 74. ht tp s://doi. org/10 .3389/
MacLean , E. L., Wilson, S. R., Martin , W. L., Davis, J. M., Nazarloo, H. P.,
& Carter, C. S. (2019). Challenges for measuring oxytocin: The blind
men and the elephant? Psychoneuroendocrinology, 107, 225– 231. uen.2019.05.018
Malin, J. L., Karberg, E., Cabrer a, N. J., Rowe, M., Cristaforo, T., &
Tamis- LeMonda, C. S. (2012). Fathertoddler communication
in low- income families: The role of paternal education and de-
pressive symptoms. Family Science, 3(3– 4), 155– 163. https://doi.
org/10.1080/19424 620. 2012.779423
Marsh, A. A. (2016). Understanding amygdala responsiveness to fearful
expressions through the lens of psychopathy and altruism. Journal
of Neuroscience Research, 94(6), 513– 525.
jnr.23 668
Mascaro, J. S., Hackett, P. D., & Rilling, J. K. (2014). Differential neural
responses to child and sexual stimuli in human fathers and non-
fathers and their hormonal correlates. Psychoneuroendocrinology,
46, 153– 163. uen.2014.04.014
Molenberghs, P., Johnson, H., Henry, J. D., & Mattingley, J. B. (2016).
Understanding the minds of others: A neuroimaging meta- analysis.
Neuroscience & Biobehavioral Reviews, 65, 276– 291. https://doi.
org/10.1016/j.neubi orev.2016.03.020
Morris, A. R., Turner, A., G ilbertson, C. H., Corner, G., Mendez, A. J., &
Saxbe, D. E. (under review). Physical touch during father- infant
interac tions is associated with paternal oxy tocin levels. Infant
Behavior and Development.
Naber, F., van Ijzendoorn, M. H., Deschamps, P., van Engeland, H., &
Bakermans- Kranenburg, M. J. (2010). Intranasal oxytocin increases
father s' observed responsiveness during play with their children: A
double- blind within- subject experiment. Psychoneuroendocrinology,
35(10), 1583– 1 586.
Norhol t, H. (2020). Revisi ting the roots of at tachment: A rev iew of the bi-
ological and psychological effects of maternal skin- to- skin contact
and carrying of full- term infants. Infant Behavior and Development,
60, 101441. h ttps :// .infbeh. 2020.101441
Noriuchi, M., Kikuchi, Y., & Senoo, A. (2008). The functional neuro-
anatomy of maternal love: Mother's response to infant's attach-
ment behaviors. Biological Psychiatry, 63, 415– 423. https://doi.
org/10.1016/j.biops ych.2007.05.018
Pancsofar, N., & Vernon- Feagans, L. (2010). Fathers' early contributions
to children's language development in families from low- income
rural communities. Early Childhood Research Quarterly, 25(4), 450–
Rilling, J. K. (2013). The neural and hormonal bases of human parental
care. Neuropsychologia, 51(4 ), 731– 747. htt ps://doi .org /10.10 16/j.
neuro psych ologia.2012.12.017
Rilling, J. K., & Mascaro, J. S. (2017). The neurobiology of fatherhood.
Current Opinion in Psychology, 15, 26– 32. https ://
Rizzolatti, G., & Cr aighero, L. (2004). The mirror- neuron system. Annual
Review of Neuroscience, 27(1), 169– 192.
annur ev.neuro.27.070203.144230
Sarka di, A., Kris tiansson, R. , Oberklaid, F., & Brem berg, S. (200 8). Fathers'
involvement and children's developmental outcomes: A systematic
review of longitudinal s tudies. Acta Paediatrica, 97(2), 153– 158. 2227.2007.00572.x
Saxbe, D. E., Edelstein, R. S., Lyden, H . A., Wardecker, B. M., Chopik, W.
J., & Moors, A. C. (2017). Fathers' prenatal testosterone decline and
synchrony with partner testosterone predicts greater postpartum
relationship investment. Hormones & Behavior, 90, 3947.
Saxbe, D., Khaled, M., Horton, K. T., & Mendez, A . J. (2019). Maternal
prenatal plasma oxytocin is positively associated with prenatal
psychological symptoms, but method of immunoassay extraction
may affec t results . Biological Psychology, 147, 107718. https://doi.
org /10.1016/j.biops ych o.2019.107718
Saxe, R., Moran, J. M., Scholz, J., & Gabrieli, J. (2006). Over lapping and
non- overlapping brain region s for theory of min d and self reflec tion
in individual subjects. Social Cognitive and Affective Neuroscience,
1(3), 229– 234.
Scatliffe, N., Casavant, S., Vittner, D., & Cong, X. (2019). Oxy tocin and
early parent- inf ant interac tions: A systematic review. International
Journal of Nursing Sciences, 6(4), 445– 453.
ij ns s. 20 19.09.00 9
Seghier, M. L. (2013). The angular gyrus: Multiple functions and mul-
tiple subdivisions. The Neuroscientist: A Review Journal Bringing
Neurobiology, Neurology and Psychiatry, 19(1), 43– 61. https://doi.
org /10.1177/1073 8 58 412 44 0596
Shulman, G. L., Pope, D. L. W., Astafiev, S. V., McAvoy, M. P., Snyder,
A. Z., & Corbetta, M. (2010). Right hemisphere dominance during
spatial selective attention and target detection occurs outside
the dorsal frontoparietal network. Journal of Neuroscience, 30(10),
3 6 4 0 3 6 5 1 . h t t p s : / / d o i . o r g / 1 0 . 1 5 2 3 / J N E U R O S C I . 4 0 8 5 - 0 9 . 2 0 1 0
Smith, S . M. (2002). Fast rob ust automated br ain extraction. Human B rain
Mapping, 17(3), 143– 155.
Smith, S. M., Jenkinson, M., Woolrich, M. W., Beckmann, C. F., Behrens,
T. E. J., Johansen- Berg, H ., Bannister, P. R., De Luca, M ., Drobnjak,
I., Flitney, D. E., Niazy, R. K., Saunders, J., Vickers, J., Zhang, Y.,
De Stefano, N., Brady, J. M., & Matthews, P. M. (2004). Advances
in functional and structural MR image analysis and implemen-
tation as FSL. NeuroImage, 23(Suppl 1), S208– S219. https://doi.
org/10.1016/j.neuro image.2004.07.051
Spunt, R . P., & Adolphs, R. (2014). Validating the why/how contrast for
functional MRI studies of theory of mind. NeuroImage, 99, 301– 311. image.2014.05.023
Spunt, R . P., & Lieberman, M . D. (2012). Dissoci ating modali ty- specific and
supramodal neural systems for action understanding. The Jour nal of
Neuroscience, 32(10), 3575– 3583. http s://doi.or g/10.1523/JNEUR
O S C I . 5 7 1 5 - 1 1 . 2 0 1 2
Storey, A. E., Walsh, C. J., Quinton, R. L., & Wynne- Edwards, K. E. (2000).
Hormonal correlates of paternal responsiveness in new and exp ect-
ant fathers. Evolution and Human Beha vior, 21(2), 79– 95. https://doi.
o r g / 1 0 . 1 0 1 6 / S 1 0 9 0 - 5 1 3 8 ( 9 9 ) 0 0 0 4 2 - 2
Storey, A. E., & Ziegler, T. E. (2016). Primate paternal care: Interactions
between biology and social experience. Hormones and Behavior, 77,
260– 271.
Swain, J. E. (2008). Baby stimuli and the parent brain: Functional neu-
roimaging of the neural substrates of parent- infant attachment.
Psychiatry, 5(8), 28– 36.
Swain, J. E., Kim, P., Spicer, J., Ho, S. S., Day ton, C. J., Elmadih, A., & Abel,
K. M. (2014). Approaching t he biology of human parental attach-
ment: Brain imaging, oxytocin and coordinated assessments of
mothers and fathers. Brain Resea rch, 1580, 78– 101. https://doi.
org/10.1016/j.brain res.2014.03.007
Szeto, A., McCabe, P. M., Nation, D. A., Tabak, B. A., Rosset ti, M. A.,
McCullough, M. E., Schneiderman, N., & Mendez, A . J. (2011).
Evaluation of enzyme immunoassay and radioimmunoassay
methods for the measurement of plasma oxytocin. Psychosomatic
Medicine, 73(5), 393– 400.
e 3 1 8 2 1 d f 0 c 2
Tabak, B. A., McCullough, M. E., Szeto, A., Mendez, A. J., & McCabe, P.
M. (2011). Oxy tocin Indexes relational distress following interper-
sonal harms in women. Psychoneuroendocrinology, 36 (1), 115– 122. uen.2010.07.004
Thesen , S., Heid, O., Mueller, E., & Schad, L. R. (20 00). Prospective ac-
quisition correction for head motion with image- based tracking for
real- time fMRI. Magnetic Resonance in Medicine, 44(3), 457– 465. 2594(20000 9)44:3<457:aid- mrm17
>;2- r
Wagner, A. D., Maril, A., Bjork, R. A ., & Schacter, D. L. (2001). Prefrontal
contributions to executive control: fMRI evidence for functional dis-
tinctions within lateral prefrontal cortex. NeuroImage, 14, 1337– 13 47.
Weime r, A. A ., Paraul t Do wds, S. J., Fabric iu s, W. V., Sch wanenfl ug el, P. J.,
& Suh, G. W. (2017). Development of constructivist theory of mind
from middle childhood to early adulthood and it s relation to social
cognition and behavior. Journal of Experimental Child Psychology,
154, 28– 45.
Weisman, O., Zagoory- Sharon, O., & Feldman, R. (2014). Oxytocin ad-
ministration, salivary testosterone, and fatherinfant social behav-
ior. Progress in Neuro- Psychopharmacology and Biological Psychiatry,
49, 4752.
Winstanley, A., & G attis, M. (2013). The Baby Care Questionnaire: A
measure of parenting principles and practices during infancy. Infant
Behavior & Development, 36(4), 762– 775.
Wittfoth- Schardt, D., Gründing, J., Witt foth, M., Lanfermann, H.,
Heinrichs, M., Domes, G., Buchheim, A., Gündel, H., & Waller, C.
(2012). Oxytocin modulates neural reactivity to children's faces
as a function of social salience. Neuropsychopharmacology, 37(8),
1799– 1807. https://doi.o rg/10 .1038/n pp. 2012.47
Wu, H., Tang, H., Ge, Y., Yang, S., Mai, X., Luo, Y. J., & Liu, C. (2017).
Object words modulate the activity of the mirror neuron system
during action imitation. Brain and Behavior, 7(11), e0084 0. https://
Yue, T., Jiang, Y., Yue, C., & Huang, X. (2017). Differential effects of oxy-
tocin on visual perspective taking for men and women. Frontiers
in Behavioral Neuroscience, 11, 228.
How to cite this article: Cardenas SI, Stoycos SA, Seller y P,
et al. Theory of mind processing in expectant fathers:
Associations with prenatal oxytocin and parental
attunement. Dev Psychobiol. 2021;00:1–19. ht t p s :// d oi .
... In direct pathways, expectant fathers influence child development through health and lifestyle factors shaping epigenetic processes or through neuroendocrine changes influencing fathers' caregiving behavior. For instance, expectant fathers' health behaviors prior to conception have been linked to altered sperm quality, fetal gestational weight, and infant birth weight (Moss & Harris, 2015), and preliminary evidence suggests that expectant fathers' prenatal neurobiology may be linked with fathers' parenting style (e.g., Cardenas et al., 2021). In indirect pathways, fathers affect child development indirectly through their influences on mothers or on the co-parenting couple relationship. ...
... For example, in a study by our research group, first-time U.S. expectant fathers with higher prenatal testosterone levels had stronger neural activation to crying sounds in both auditory processing and mentalizing network regions , and subsequently reported less effective and positive parenting at 6 months postpartum . Consistent with this, expectant fathers who showed higher neural activation on a standardized mentalizing task displayed higher prenatal oxytocin levels and endorsed a more attunementoriented parenting philosophy after their child's birth (Cardenas et al., 2021). In another prospective, longitudinal study, first-time Dutch fathers displayed similar neural responses from the prenatal to the postpartum period when watching videos of infants being put in threatening situations (e.g., a baby in a stroller being pushed into a river) compared to neutral situations (e.g., a baby in a stationary stroller; van 't Veer et al., 2019), and fathers' postpartum neural responses to infant threat were associated with their parental protectiveness in daily life. ...
Full-text available
Extensive research has established that fathers’ engagement in parenting benefits children, but few studies have described how fathers contribute to child development even before birth. In this article, we consider both direct and indirect pathways through which expectant fathers shape child development during the prenatal period. Regarding direct pathways, we review work on expectant fathers’ contributions to child development through genetic and epigenetic processes, as well as neuroendocrine mechanisms. Regarding indirect pathways, we outline ways in which expectant fathers indirectly influence child development through the couple relationship. In so doing, we seek to provide a foundation from which to formulate future lines of inquiry on the role of expectant fathers in child development. This research can inform clinical interventions and policies geared toward improving the early caregiving environment and child development.
... Data for this study were drawn from a larger longitudinal study on the transition to parenthood that took place in a large city in California (Cardenas et al., 2021;Khaled et al., 2020Khaled et al., , 2021Khoddam et al., 2020;Saxbe et al., 2019;Saxbe et al., 2018aSaxbe et al., , 2018b. Couples were recruited during pregnancy through fliers posted in obstetricians' offices, community health clinics, and social media. ...
... For example, parenting experience can affect neural responses to infant stimuli, suggesting that time with infants can remodel the brain (Abraham et al., 2014). Moreover, the parenting brain and biology literatures points to interesting differences between fathers and mothers that warrant further investigation and support the importance for specifically studying fathers' engagement in parenting in addition to mothers (Cardenas et al., 2021;Seifritz et al., 2003). ...
Full-text available
Paid family leave may mitigate stress and health challenges across the transition to parenthood. The current study examined whether paid paternity leave is associated with first-time parents' trajectories of depression, stress, and sleep from the prenatal to postpartum periods. Expectant parents (72 couples) reported on their depressive symptoms, perceived stress, and daytime fatigue during mid-to-late pregnancy and then again at six months postpartum. At one year postpartum, fathers reported on any paid or unpaid leave taken following their child's birth. We used a repeated-measures design to compare couples in which the father either did or did not access paid paternity leave. When fathers took paid paternity leave, their partners' stress and depressive symptoms showed smaller prenatal to postpartum increases than mothers whose partners did not take paid leave. Similarly, fathers who took paid paternity leave, compared to those who did not, reported smaller prenatal to postpartum increases in stress and daytime fatigue. These results remained largely unchanged when controlling for the length of fathers' leave. The study's longitudinal, within-subject design allows us to examine parents' mental health relative to their own prenatal baseline, helping to account for some of the pre-existing differences between fathers who did and did not take paid paternity leave. The results suggest that paid paternity leave may be associated with greater well-being across the transition to parenthood for both fathers and mothers.
... Horrell et al. (2021) reviewed the extant literature connecting the transition to fatherhood with cognitive and emotional changes that are associated with a burst of neuroplasticity thought to promote infant care by fathers. Cardenas et al. (2021) found that higher levels of prenatal neural activation on a theory of mind task and oxytocin were related to postnatal parental-specific theory of mind, suggesting that prenatal theory of mind may be an important precursor to later fathering. ...
... These changes were associated with gestational age and the men's perception of parenting (Diaz-Rojas et al., 2021). A recent prenatal fMRI study reported greater activations in regions that support mentalization, including STS and dmPFC, during a "Theory of Mind" task in a sample of 39 expected fathers in the USA, which were associated with fathers' parenting believes about attuned parenting at three months postpartum (Cardenas et al., 2021). A longitudinal fMRI study of processing threat to infants examined the neural basis for protective parenting before and after the birth of Dutch fathers' first child. ...
Full-text available
With the growing involvement of fathers in childrearing and the application of neuroscientific tools to research on parenting, there is a need to understand how a father's brain and neurohormonal systems accommodate the transition to parenthood and how such neurobiological changes impact children's mental health, sociality, and family functioning. In this paper, we present a theoretical model on the human father’s brain and the neural adaptations that take place when fathers assume an involved role. The neurobiology of fatherhood shows great variability across individuals, societies, and cultures and is shaped to a great extent by bottom-up caregiving experiences and the amount of childrearing responsibilities. Mechanisms of motherfather coparental brain coordination and hormonal correlates of paternal behavior are detailed. Adaptations in the father’s brain during pregnancy and across the postpartum year carry long-term implications for children's emotion regulation, stress management, and symptom formation. We propose a new conceptual model of HEALthy Father Brain that describes how a father’s brain serves as a source of resilience in the context of family adversity and its capacity to “heal”, protect, and foster social brain maturation and functionality in family members via paternal sensitivity, attunement, and support, which, in turn, promote child development and healthy family functioning. Father’s brain provides a unique model on neural plasticity as sustained by committed acts of caregiving, thereby affording a novel perspective on the brain basis of human affiliation.
We compared self-reported parenting beliefs about caring for infants with observed parenting behaviours during play interactions between 32 parents and their infants. We measured parenting beliefs about the value of attunement and structure in caring for infants using the Baby Care Questionnaire (BCQ) (Winstanley & Gattis, 2013; Winstanley, Sperotto, Putnick, Cherian, Bornstein & Gattis, 2014). We used a micro-coding approach to distinguish between responsive parenting behaviours (maintaining infant attention) and demanding parenting behaviours (introducing or redirecting infant attention) (Landry, Garner, Swank & Baldwin, 1996). Attunement beliefs were positively related to responsive parenting behaviours and negatively related to demanding parenting behaviours. Structure beliefs were weakly related to demanding parenting behaviours. These results are an important first step toward identifying relations between self-reported parenting beliefs about attunement and structure and observed parenting behaviours.
Full-text available
Women that become mothers face notable physiological adaptations during this life-period. Neuroimaging studies of the last decade have provided grounded evidence that women's brains structurally change across the transition into motherhood. The characterization of this brain remodeling is currently in its early years of research. The current article reviews this scientific field by focusing on our longitudinal (pre-to-post pregnancy) Magnetic Resonance Imaging (MRI) studies in first-time parents and other longitudinal and cross-sectional studies of parents. We present the questions that are currently being answered by the parental brain literature and point out those that have not yet been explored. We also highlight potential confounding variables that need to be considered when analyzing and interpreting brain changes observed during motherhood.
Full-text available
Expectant parents' responses to infant cry may indicate future risk and resiliency in the parent-child relationship. Most studies of parental reactivity to infant cry have focused on mothers, and few studies have focused on expectant fathers, although fathers make important contributions to parenting. Additionally, although different responses to infant cry (behavioral, psychological, and neural) are hypothesized to track together, few studies have analyzed them concurrently. The current investigation aimed to address these gaps by characterizing multimodal responses to infant cry within expectant fathers and testing whether prenatal testosterone moderates these responses. Expectant fathers responded to infant cry versus frequency-matched white noise with increased activation in bilateral areas of the temporal lobe involved in processing speech sounds and social and emotional stimuli. Handgrip force, which has been used to measure parents' reactivity to cry sounds in previous studies, did not differentiate cry from white noise within this sample. Expectant fathers with higher prenatal testosterone showed greater activation in the supramarginal gyrus, left occipital lobe and precuneus cortex to cry sounds. Expectant fathers appear to interpret and process infant cry as a meaningful speech sound and social cue, and testosterone may play a role in expectant fathers' response to infant cry.
Full-text available
Pregnancy and the transition to parenthood is an important period marked by dramatic neurobiological and psychosocial changes that may have implications for the health of women and offspring. Although human and non-human animal research suggests the brain undergoes alterations during the peripartum period, these changes are poorly understood. Here, we review existing research, particularly human neuroimaging and psychophysiological research, to examine changes in brain structure and function during the peripartum period and discuss potential implications for the health of women and offspring. First, we discuss the potential causes of these changes across pregnancy, including physiological and psychosocial factors. Next, we discuss the evidence for structural and functional changes in the brain during pregnancy and into the postpartum period, noting the need for research conducted prospectively across human pregnancy. Finally, we propose potential models of individual differences in peripartum neurobiological changes (i.e., hypo-response, typical response, hyper-response) and emphasize the need to consider trajectories of change in addition to pre-existing factors that may predict maternal adjustment to parenthood. We suggest that the consideration of individual differences in neurobiological trajectories across pregnancy may contribute to a better understanding of risk for negative health and behavior outcomes for women and offspring.
Full-text available
Objectives: Social relationships throughout lifespan are critical for health and wellbeing. Oxytocin, often called the 'hormone of attachment' has been suggested as playing an important role in early-life nurturing and resulting social bonding. The objective of this paper is to synthesize the associations between oxytocin levels and interactions between infants and parents that may trigger oxytocin release, and in turn facilitate attachments. Methods: A comprehensive cross-disciplinary systematic search was completed using electronic databases. The inclusion criteria included studies that focused on mother-infant and father-infant interaction and measured both baseline and post-interaction oxytocin levels. Results: Seventeen studies were included in the final systematic review. The reviewed studies used mother-infant and/or father-infant play and skin-to-skin contact between maternal-infant and paternal-infant dyads to examine the oxytocin role in early life bonding and parenting processes. Studies showed a positive correlation between parent-infant contact and oxytocin levels in infancy period. Increased maternal oxytocin levels were significantly related to more affectionate contact behaviors in mothers following mother-infant contact, synchrony, and engagement. Meanwhile, increased paternal oxytocin levels were found to be related to more stimulatory contact behaviors in fathers following father-infant contact. Oxytocin levels significantly increased in infants, mothers and fathers during skin-to-skin contact and parents with higher oxytocin levels exhibited more synchrony and responsiveness in their infant interactions. Conclusion: The review suggests that oxytocin plays an important role in the development of attachment between infants and parents through early contact and interaction. The complexities of oxytocinergic mechanisms are rooted in neurobiological, genetic, and social factors.
Full-text available
Social support plays a vital role in physical and mental well-being. The neuropeptide hormone oxytocin (OXT) has been implicated in modulating pair-bonding and affiliative behaviors, but whether OXT contributes to the analgesic effects of a romantic partner's touch remains elusive. In the present randomized placebo-controlled, between-group, functional magnetic resonance imaging study involving 194 healthy volunteers (97 heterosexual couples), we tested the effects of intranasal OXT (24 IU) on handholding as a common mode of expressing emotional support in romantic couples. We scanned the subjects while brief electric shocks were administered. The subjects assumed that they received social support from either their romantic partner or an unfamiliar person. Unbeknown to the subject, in the partner and stranger support conditions, the same male experimenter always held the subject's left hand. Partner support was most effective in reducing the unpleasantness of electric shocks, and OXT further attenuated the unpleasantness across conditions. On the neural level, OXT significantly augmented the beneficial effects of partner support, as evidenced by a stronger decrease of neural responses to shocks in the anterior insula (AI), a stronger activity increase in the middle frontal gyrus (MFG), and a strengthened functional coupling between the AI and MFG. Our results support the notion that OXT specifically modulates the beneficial effects of social support in romantic couples by concomitantly reducing pain-associated activity and increasing activity linked to cognitive control and pain inhibition. We hypothesize that impaired OXT signaling may contribute to the experience of a lack of partner support.
Full-text available
Why do fathers matter? Recent conceptual and theoretical advances regarding father–child relationships have demonstrated that fathers affect children's outcomes both directly and indirectly. To attain a complete developmental account of the ecologically rich contexts of child development, in this article, we recommend best practices regarding the conceptualization and assessment of father–child relationships that reflect contemporary family life. We also discuss conceptual and measurement issues pertaining to father–child relationships in different family configurations, including those with resident and nonresident fathers. We conclude with recommendations that can help developmental researchers advance our understanding of fathering, parenting, and children's development.
Full-text available
Oxytocin (OT) concentration in the blood is considered to be a marker of its action in the brain. However, two problems have emerged when measuring OT level in the blood. First, it is unclear whether different methods of assessment lead to similar OT values. Second, it is unclear if plasma OT concentrations is informative on what OT does in the brain. To clarify these issues, we collected cerebrospinal fluid (CSF) from the brain ventricle of 25 patients during surgery to compare with plasma OT after simultaneous blood withdrawal. Additionally, we collected 12 CSF and blood samples from non-human primates while awake or under anaesthesia. We used four methods to assay OT concentrations: Commercial EIA with/without extraction, laboratory developed EIA with filtration and RIA with extraction. Three of these methods showed a positive correlation between plasma and CSF OT, suggesting a link between plasma and central OT, at least under specific testing conditions. However, none of the methods correlated to each other. Our results show major disagreements among methods used here to measure peripheral and brain OT and therefore they call for more caution when plasma OT is taken as a marker of central OT.
Introduction Physical connection, particularly parent-to-infant touch, is critical for the well-being of infants and may support the development of the parent-infant bond. Physical touch has also been found to stimulate oxytocin levels. This study tested whether fathers' micro-coded touch behaviors during parent-child interaction predicted their subsequent oxytocin levels. We also compared two widely-used methods of oxytocin immunoassay that have been found to yield discrepant results in past studies. Methods Among 45 fathers and their six-month-old infants, we micro-coded paternal physical touch at 1/10 s intervals during a laboratory-based free-play interaction. Paternal oxytocin was measured via blood plasma and was processed both with and without the extraction step prior to immunoassay so that results from the two methods could be compared. Results Unextracted and extracted oxytocin were moderately correlated within our sample. Fathers who engaged in more playful proprioceptive touch showed higher levels of both unextracted and extracted oxytocin. Gentle affectionate touch and functional proprioceptive touch predicted higher unextracted but not extracted oxytocin levels. Fathers who did not engage in physical touch showed lower levels of both unextracted and extracted oxytocin. Conclusion Results are consistent with previous work showing that physical touch, particularly playful proprioceptive touch, is associated with higher oxytocin levels in fathers. These results replicate previous research using unextracted oxytocin measurement, and extend this work, showing that many but not all associations hold when using the more rigorous method of extraction when measuring oxytocin.
During the early period of hypothesis building and empirical testing of attachment theory, a major emphasis was placed on mother-infant physical contact. In spite of this, mother-infant contact has received scant attention amongst attachment and child development researchers in the past decades. Here, a brief theoretical framework for mother-infant contact is presented, drawing on animal studies as well as human studies of preterm infants and neonates. Salient mechanisms may include an extended sensitive period during early infancy, requiring specific somatosensory stimuli for bio-behavioral homeorhesis; oxytocinergic and epigenetic pathways; kinesthetic stimuli and face-to-face proximity allowing for increased social interaction. Studies of extended human mother-full-term infant physical contact have demonstrated positive effects in multiple domains. For infants, these include sleep organization, temperature and heart rate regulation, behavioral response, crying/colic, socio-emotional development, attachment quality, speech development opportunities and mother-child interactions. For mothers, studies demonstrate improved depressive symptomatology, physiological stress regulation, contingent responsivity, breastfeeding and mother-child interactions. Parent-infant attachment quality has gained prominence as a trauma-resilience factor as well as a predictor of adult physical health. The potential role of mother-infant contact as an attachment promoting intervention as well as future research subjects are discussed. Current evidence supports the original attachment research that early maternal touch provision may influence infant socio-emotional development and attachment quality, with positive implications for mother-child relationship functioning.
We examined associations between prenatal plasma oxytocin levels and depressive symptoms, state anxiety, and pregnancy anxiety in 75 women who visited the laboratory with their partners during mid-to-late pregnancy and engaged in relationship discussion tasks prior to a blood draw. Given controversy in the literature regarding oxytocin measurement, we compared two widely-used immunoassay approaches (with and without extraction prior to immunoassay). Levels of immunoreactive oxytocin measured with and without extraction were not correlated with each other. However, both extracted and unextracted oxytocin were positively associated with women's prenatal depressive symptoms in a model that controlled for pregnancy stage and body mass index. Only unextracted oxytocin was associated with state anxiety and pregnancy-specific anxiety. In summary, elevated plasma oxytocin levels in expectant mothers might indicate risk for mental health symptoms during the prenatal period, but results for anxiety are mixed and appear to depend on the immunoassay approach employed.
Since its discovery more than a century ago, oxytocin has become one of the most intensively studied molecules in behavioral biology. In the last five years, Psychoneuroendocrinology has published more than 500 articles with oxytocin in the title, with many of these articles including measures of endogenous oxytocin concentrations. Despite longstanding interest, methods of measuring endogenous oxytocin are still in active development. The widely varying oxytocin concentrations detected by different approaches to measurement - and lack of correlation among these techniques - has led to controversy and confusion. We identify features of oxytocin that may help to explain why various approaches may be differentially sensitive to diverse conformational states of the oxytocin molecule. We propose that discrepancies in data generated by different methods of measurement are not necessarily an indicator that some methods are valid whereas others are not. Rather, we propose that current challenges in the measurement of oxytocin may be analogous to the parable of the blind men and the elephant, with different methods of sample preparation and measurement being sensitive to different states in which the oxytocin molecule can exist.