Oxytocin Enhances the Encoding of Positive Social

Memories in Humans
Adam J. Guastella, Philip B. Mitchell, and Frosso Mathews
Background: In nonhuman mammals, oxytocin has a critical role in social recognition and the development of long-term bonds. There has
been limited research evaluating effects of oxytocin on the encoding and recognition of faces in humans.

Methods: In a double-blind, randomized, placebo-controlled, between-subject design, we administered oxytocin (24 IU) or a placebo to 69
healthy human male volunteers and then presented 36 happy, angry, or neutral human faces. Participants returned the following day to
make “remember,” “know,” or “new” judgments for a mix of 72 new and previously seen faces.

Results: Oxytocin-administered participants were more likely to make remember and know judgments for previously seen happy faces
compared with angry and neutral human faces. In contrast, oxytocin did not influence judgments for faces that had not been presented
previously.

Conclusions: This study shows that the administration of oxytocin to male humans enhances the encoding of positive social information
to make it more memorable. Results suggest that oxytocin could enhance social approach, intimacy, and bonding in male humans by
strengthening encoding to make the recall of positive social information more likely.

Key Words: Emotion, face recognition, oxytocin, peptide, social research shows important differences between remember (“rec-
cognition ollection of detail”) and know (“feels familiar”) judgments (15),
and these seem to be based on discernable biological processes
(16). We selected three types of facial expressions for study:

T
he benefits of maintaining supportive social relationships
are well documented (1). Researchers suggest a role for the
neuropeptide oxytocin (OT) in this process by enhancing
social-reward and reducing stress and social-threat responding
(2,3). Animal studies provide strong evidence implicating OT in
social behaviour. For instance, mice given an injection of OT into
the medial-amygdala display improved recognition for peers
(4,5). Although OT knockout mice fail to show evidence of
social-recognition (5), administration of OT to these animals
restores recognition (4). These effects are found when OT is
administered before social encounters but not after, suggest-
ing that OT enhances encoding rather than retrieval (4).
Effects in animals also seem specific to rewarding social
stimuli (6). Oxytocin has amnesic effects on the consolidation
of nonsocial stimuli (7) and reduces social aggression and
threat responses (6).
Social recognition is essential for developing long-term bonds
(8). We could not locate any evaluation of OT influences on
recognition memory for faces in humans. Human research
suggests that OT impairs or does not influence memory for
nonsocial information, such as word lists or general emotional
information (9 –11). There is evidence, however, that OT en-
hances the perception of faces in males. OT increases gaze to the
eye region of human faces (12), and improves the identification
of emotion from the eyes of others (13). The aim of this study
was to evaluate whether OT given before encoding would
enhance face recognition memory long term. We used the
remember/know paradigm (14) because a substantial body of
From the Brain & Mind Research Institute (AJG), University of Sydney, School
of Psychology (AJG, FM), and School of Psychiatry (PBM), University of
New South Wales, Sydney, Australia.
Address reprint requests to Adam Guastella, Ph.D., Brain & Mind Research Insti-
tute, University of Sydney, NSW, Australia; E-mail: aguastella@med.
usyd.edu.au.
Received November 21, 2007; revised February 12, 2008; accepted February
13, 2008.
happy, angry, and neutral. Researchers suggest that OT may
produce effects, in part, by enhancing the reward value for
positive social stimulus (2,17) while decreasing responding to
aggressive social stimulus (3,11). It was hypothesized, therefore,
that OT would enhance encoding and subsequent recognition
for happy faces and reduce encoding and subsequent recogni-
tion for angry faces.
Methods and Materials
Sixty-nine healthy young adult male students1 from the
University of New South Wales (UNSW; aged 18 –30 years, M ⫽
19.98, SD ⫽ 2.27) were randomly assigned in a double-blind
manner to receive either 24 international units (IU) of OT (n ⫽
35; Novartis, Basel, Switzerland) or a placebo (n ⫽ 34). Nasal
sprays were developed by a compounding chemist (four puffs
per nostril, each with 3 IU), with an identical placebo containing
all ingredients except the active OT. Exclusion criteria included
self-reported symptoms of depression, bipolar, panic, and psy-
chotic disorders; substance dependence; epilepsy; and traumatic
brain injury. Exclusion was determined from a brief self-report
interview by a trained experimenter. Participants were instructed to
abstain from alcohol and caffeine on the day of the procedure and
food and drink (except water) 2 hours before drug administration.
After a study description, written consent was obtained, and partic-
ipants were told they could withdraw at any time. Ethical approval
was provided by the UNSW Ethics Committee (06074).
The Positive and Negative Affect Scale (PANAS) (18) was used
to track mood changes in positive and negative affect states. Face
stimuli consisted of 72 different male and female human black
and white happy, angry, and neutral face expressions (24 of each
expression). Seventy faces were taken from the Karolinska-
Directed Emotional Faces database (19). Two additional faces
1
It is standard practice in the literature (12,13) to exclude females because
of variation in estrogen levels across the menstrual cycle and ethical
considerations relating to giving OT during pregnancy (32).

0006-3223/08/$34.00 BIOL PSYCHIATRY 2008;64:256 –258

doi:10.1016/j.biopsych.2008.02.008 © 2008 Society of Biological Psychiatry
A.J. Guastella et al.
Table 1. Percentage of Raw Remember, Know, and New Judgments Made for Old and New Face Expressions
Mean (SD)
Old Stimuli
Remember (%) Know (%)
Angry Faces
OT M 58 (.23) 23 (.15)
Placebo M 55 (.18) 29 (.17)
Happy Faces
OT M 55 (.22) 30 (.18)
Placebo M 43 (.17) 31 (.15)
Neutral Faces
OT M 43 (.18) 29 (.18)
Placebo M 46 (.19) 31 (.15)
OT ⫽ oxytocin.
were edited and matched in appearance from the NimStim
Face-Stimulus database (20). No face identity was used more
than once. Twelve happy, twelve angry, and twelve neutral face
expressions were randomly selected to serve in Set A or Set B.
Sets were counterbalanced as “old” (presented at study and test)
or “new” (presented only at test) for the remember/know
procedure.
Participants completed the PANAS, the brief interview, and
then received the nasal spray. Forty-five minutes later (12,13),
participants completed the PANAS and viewed 36 faces (Set A or
Set B) for 2 sec on a standard 17-inch computer monitor.
Participants were told that we were interested in how OT
influenced reactions to social information and asked to make
ratings on a computer of how critical (1 ⫽ extremely critical to
9 ⫽ extremely accepting) or trustworthy (1 ⫽ not at all to 9 ⫽
extremely) each face appeared. Participants returned the follow-
ing day for a surprise memory test consisting of 72 randomized
faces from Set A and B. Participants were asked to press the R
(Remember) key if they could recollect specific details about a
face from the study session (such as an aspect of the face’s
physical appearance), the K (Know) key if the face felt familiar
but they could not recollect specific details, and the N (New) key
if they believed the face was not presented at study. Participants
completed self-reports (PANAS, possible drug side effects, beliefs
about drug assignment) and were debriefed.
Results
Eight participants were excluded due to equipment malfunc-
tion or failure to attend sessions, leaving 31 OT and 30 placebo
participants with no age difference [t (59) ⫽ .99, p ⫽ .33]. t tests
and chi-square analysis evaluating differences between drug
groups showed there were no differences in positive or negative
affect ratings at any session [largest t (59) ⫽ –1.58, p ⫽ .12], no
differences across critical and trustworthiness ratings of faces,2
and no differences in the number or type of side effects reported3
or beliefs about drug allocation (p ⬎ .3). Table 1 presents the raw
percentages (before transformation) of remember, know, and
new responses for “old” and “new” face stimuli. t tests run on
2
A trend from the t tests run indicated OT participants could have rated
happy faces as more trustworthy than placebo participants [t (59) ⫽
1.85, p ⫽ .07; OT M ⫽ 6.13; SD ⫽ 1.05; placebo M ⫽ 5.61; SD ⫽ 1.12].
3
Side effects reported by OT participants included relaxation or tiredness
(n ⫽ 8), attentiveness (n ⫽ 2), headaches (n ⫽ 2), and blurred vision
(n ⫽ 1). Placebo participants reported relaxation or tiredness (n ⫽
10), attentiveness (n ⫽ 1), and irritability (n ⫽ 1).
BIOL PSYCHIATRY 2008;64:256 –258 257
New Stimuli
New (%) Remember (%) Know (%) New (%)
19 (.17) 07 (.09) 20 (.14) 73 (.18)
16 (.12) 04 (.05) 21 (.15) 75 (.17)
15 (.13) 04 (.07) 21 (.16) 75 (.19)
26 (.16) 03 (.05) 17 (.12) 80 (.15)
28 (.16) 06 (.09) 19 (.13) 75 (.16)
24 (.17) 07 (.10) 21 (.12) 72 (.16)
judgments for “new” happy, angry, and neutral faces showed no
differences between drug groups4 [largest t (59) ⫽ 1.50, p ⫽ .14].
Scores for “old” stimuli were subjected to the independence
of redundancy assumption (16,21): Independent Remember ⫽
(Hit Remember – False Alarm Remember); Independent Know ⫽
Know/(1 – Remember). A 2 [Drug (OT, Placebo)] ⫻ 3 [Face
Expression (Happy, Angry, Neutral)] ⫻ 2 [Judgment (Indepen-
dent Remember, Independent Know)] repeated-measures multi-
variate analysis of variance was run on responses to “old” face
stimuli. Results indicated a main effect of Judgment [F (1,59) ⫽
18.55, p ⬍ .001], showing a reported higher percentage of
independent know judgments than independent remember judg-
ments. There was no main effect of Drug [F (1,59) ⫽ .01, p ⫽ .95],
but a main effect of Face Expression [F (2,58) ⫽ 6.18, p ⫽ .004],
which was better accounted for by a Face Expression by Drug
interaction [F (2,58) ⫽ 7.06, p ⫽ .002, (see Figure 1). Follow-up
repeated-measure analyses of variance for each judgment
showed that OT participants made more independent remember
judgments for happy faces5 [F (2,58) ⫽ 2.71, p ⫽ .07]. An
interaction was also found for independent know judgments
[F (2,58) ⫽ 5.65, p ⫽ .006], showing that, in contrast to placebo,
OT participants displayed enhanced independent know re-
sponding to happy faces over angry and neutral faces. Interac-
tions between Judgment and Face Expression [F (2,58) ⫽ 1.69,
p ⫽ .19], Judgment and Drug [F (1,59) ⫽ .92, p ⫽ .34], and Face
Expression, Judgment, and Drug [F (2,58) ⫽ .67, p ⫽ .51] were
nonsignificant.
Discussion
This is the first study in humans to show that OT strengthens the
encoding of positive social stimuli to make this information more
memorable. Participants administered OT at study displayed im-
proved “remember” responding for happy faces and a bias in
“know” responding for happy over angry and neutral faces. In
support of previous research (12,13), there was no clear influence of
OT on subjective self-reports, such as mood, drug-identification,
and face-appraisal ratings. Findings support evolutionary-biological
perspectives (11) that emphasize OT’s role in enhancing processing
for cues important for bonding, while reducing the impact of
socially aversive cues. Our findings indicate OT may facilitate
4
A 2 [Drug] ⫻ 3 [Face Expression)] ⫻ 2 [Judgment Type] repeated-
measures multivariate of analysis was run on responses to “new” face
stimuli. No significant main effects or interactions were found.
5
When analysis was rerun using raw score remember judgments, the
effect was stronger [F (2,58) ⫽ 4.18, p ⫽ .02], showing OT enhanced
memory for positive faces over placebo [t (59) ⫽ 2.38, p ⫽ .02].
www.sobp.org/journal
258 BIOL PSYCHIATRY 2008;64:256 –258
Figure 1. Percentages of independent remember and know judgments
across happy, angry, and neutral faces. Error bars represent standard error.
Rem ⫽ independent remember judgment; for independent remember
judgments, analysis indicated oxytocin enhancement of remember re-
sponding to happy faces [t(59) ⫽ 2.11, p ⫽ .04].
intimacy and bonding in male humans by making the encoding of
positive social cues during interactions more memorable.
We propose two biological explanations. First, animal research
shows that activation of OT receptors within the central amygdala
reduces fear (22). This has led to speculation (23) that OT reduces
social-threat detection and associated memory consolidation in
humans. Oxytocin may reduce the salience of threatening-angry
faces (24), allowing for enhanced processing of other salient cues
(i.e., positive social cues). Second, OT may directly enhance pro-
cessing for positive social-cues by activating social-reward neural
networks (25–27). The amygdala has rich links to reward-value
pathways (28) that are involved in processing romantic and attrac-
tive faces (29), as well as encoding and preference learning for
positive faces (30). Further, direct eye gaze activates reward-value
pathways (31), and OT enhances eye gaze (12). Oxytocin may
enhance reward-value for positive social cues to increase salience
and encoding. Animal OT data shows the importance of reward
pathways to bonding (8), but human OT research is needed. Taken
together, these data suggest that OT may benefit psychiatric disor-
ders characterized by social-bonding deficits (i.e., autism) and
social-threat responses (i.e., social anxiety disorder) by strengthen-
ing the encoding of positive social cues in everyday interactions.
Research is also required extending these findings to females,
particularly given implications for mother-infant bonding.
We thank Dr. Daniel Hermens, Dr. Gavan McNally, and
Professor Mark Dadds for helpful comments on this article, as
well as Dr. Kevin Bird for statistical advice. We thank Siobhan
Kelly and Caitlin Van Slooten for assistance with data collection.
Dr. Guastella, Dr. Mitchell, and Miss Mathews report no
biomedical financial interests or potential conflicts of interest.
1. Uchino B (2006): Social support and health: A review of physiological
processes potentially underlying links to disease outcomes. J Behav Med
29:377–387.
2. Uvnas-Moberg K (1998): Oxytocin may mediate the benefits of posi-
tive social interaction and emotions. Psychoneuroendocrinology 23:
819 – 835.
3. Bartz JA, Hollander E (2006): The neuroscience of affiliation: Forging
links between basic and clinical research on neuropeptides and social
behavior. Horm Behav 50:518 –528.
4. Ferguson JN, Aldag JM, Insel TR, Young LJ (2001): Oxytocin in the medial
amygdala is essential for social recognition in the mouse. J Neurosci
21:8278 – 8285.
www.sobp.org/journal
A.J. Guastella et al.
5. Young LJ (2002): The neurobiology of social recognition, approach, and
avoidance. Biol Psychiatry 51:18 –26.
6. Lim MM, Young L (2006): Neuropeptidergic regulation of affiliative be-
havior and social bonding in animals. Horm Behav 50:506 –517.
7. de Oliveira LF, Camboim C, Diehl F, Consiglia AR, Quillfeldt JA (2007):
Glucocorticoid-mediated effects of systemic oxytocin upon memory
retrieval. Neurobiol Learn Mem 87:67–71.
8. Young LJ, Wang Z (2004): The neurobiology of pair bonding. Nat Neuro-
sci 7:1048 –1054.
9. Pitman RK, Orr SP, Lasko NB (1993): Effects of intranasal vasopressin and
oxytocin on physiologic responding during personal combat imagery in
Vietnam veterans with posttraumatic stress disorder. Psychiatry Res 48:
107–117.
10. Bruins J, Hijman R, Van Ree JM (1992): Effect of a single dose of des-
glycinamide-[Arg8]vasopressin or oxytocin on cognitive processes in
young healthy subjects. Peptides 13:461– 468.
11. Heinrichs M, Meinlschmidt G, Wippich W, Ehlert U, Hellhammer DH
(2004): Selective amnesic effects of oxytocin on human memory. Physiol
Behav 83:31–38.
12. Guastella AJ, Mitchell PB, Dadds MR (2008): Oxytocin increases gaze to
the eye-region of human faces. Biol Psychiatry 63:3–5.
13. Domes G, Heinrichs M, Michel A, Berger C, Herpertz S (2007): Oxytocin
improves “mind-reading” in humans. Biol Psychiatry 61:731–733.
14. Rajaram S (1993): Remembering and knowing: two means of access to
the personal past. Mem Cogn 21:89 –102.
15. Diana RA, Reder LM, Arndt J, Park H (2006): Models of recognition: A review
of arguments in favor of a dual-process account. Psychon Bull Rev 13:1–21.
16. Sharot T, Delgado M, Phelps E (2004): How emotion enhances the feel-
ing of remembering. Nat Neurosci 2:1376 –1380.
17. Depue RA, Morrone-Strupinsky JV (2005): A neurobehavioral model of
affiliative bonding: Implications for conceptualizing a human trait of
affiliation. Behav Brain Sci 28:313–395.
18. Watson D, Clark LA, Tellegen A (1988): Development and validation of
brief measures of positive and negative affect: The PANAS scales. J Pers
Social Psychol 54:1063–1070.
19. Lundqvist D, Flykt A, Ohman A (1988): The Karolinska Directed Emo-
tional Faces—KDEF. Stockholm: Karolinska Insitutet, Department of
Clinical Neuroscience, Psychology Section.
20. Tottenham N, Tanaka JW, Leon AC, McCarry T, Nurse M, Hare TA, et al.
(2006): NimStim set of facial expressions. Boston: The MacArthur Foun-
dation Research Network on Early Experience and Brain Development.
21. Yonelinas AP, Jacoby LL (1995): The relation between remembering and
knowing as bases for recognition: Effect of size congruency. J Mem Lang
34:622– 643.
22. Hubber D, Veinante P, Stoop R (2005): Vasopressin and oxytocin excite
distinct neuronal populations in the central amygdala. Science 308:245–
248.
23. Heinrichs M, Gaab J (2007): Neuroendocrine mechanisms of stress and
social interaction: Implications for mental disorders. Curr Opin Psychiatry
20:158 –162.
24. Fox E, Lester V, Russo R, Bowles RJ, Pichler A, Dutton K (2000): Facial
expressions of emotion: Are angry faces detected more efficiently?
Cogn Emotion 14:61–92.
25. Cox SM, Andrade A, Johnsrude IS (2005): Learning to like: A role for human
orbitofrontal cortex in conditioned reward. J Neurosci 25:2733–2740.
26. Balleine BW, Delgado MR, Hikosaka O (2007): The role of the dorsal
striatum in reward and decision-making. J Neurosci 27:8161– 8165.
27. Wickens JR, Horvitz JC, Costa RM, Killcross S (2007): Dopaminergic
mechanisms in actions and habits. J Neurosci 27:8181– 8183.
28. Senior C (2003): Beauty in the brain of the beholder. Neuron 38:525–528.
29. Aron A, Fisher HE, Mashek DJ, Strong G, Li HF, Brown LL (2005): Reward,
motivation and emotion systems associated with early-stage intense
romantic love. J Neurophysiol 94:327–337.
30. Bray S, O’Doherty J (2007): Neural coding of reward-prediction error
signals during classical conditioning with attractive faces. J Neuro-
physiol 97:3036 –3045.
31. Kampe KK, Frith CD, Dolan RJ, Frith U (2002): Reward value of attractive-
ness and gaze. Nature 413:589.
32. Campbell A (2008): Attachment, aggression, and affiliation: The role of
oxytocin in female social behaviour. Biol Psychol 77:1–10.