Content uploaded by Vanessa LoBue
Author content
All content in this area was uploaded by Vanessa LoBue on Mar 11, 2018
Content may be subject to copyright.
Developmental Science 12:1 (2009), pp 201–207 DOI: 10.1111/j.1467-7687.2008.00753.x
© 2008 The Authors. Journal compilation © 2008 Blackwell Publishing Ltd, 9600 Garsington Road, Oxford OX4 2DQ, UK and
350 Main Street, Malden, MA 02148, USA.
Blackwell Publishing Ltd
PAPER
The narrow fellow in the grass: human infants associate snakes
and fear
Judy S. DeLoache and Vanessa LoBue
Department of Psychology, University of Virginia, USA
Abstract
Why are snakes such a common target of fear? One current view is that snake fear is one of several innate fears that emerge
spontaneously. Another is that humans have an evolved predisposition to learn to fear snakes. In the first study reported here,
9- to 10-month-old infants showed no differential spontaneous reaction to films of snakes versus other animals. In the second
study, 7- to 18-month-old infants associated snakes with fear: As predicted, they looked longer at films of snakes while listening
to a frightened human voice than while listening to a happy voice. In the third study, infants did not look differentially to still
photos of snakes and other animals, indicating that movement is crucial to infants’ association of snakes with fear. These results
offer support for the view that humans have a natural tendency to selectively associate snakes with fear.
Introduction
Snakes have long served as representations of danger
and evil, from paintings of Eve with the Serpent in the
Garden of Eden to Emily Dickinson’s poem in which the
sight of a ‘narrow fellow in the grass’ causes ‘a tighter
breathing and zero at the bone’. The ubiquity of such
images is not surprising, given the fact that snakes
constitute one of the most common objects of intense fears
and phobias (Agras, Sylvester & Oliveau, 1969; Fredrikson,
Annas, Rischer & Wik, 1996; King, 1997). Fearful reactions
to snakes have also been reported for a variety of non-
human primates (e.g. Joslin, Fletcher & Emlin, 1964;
Rumbaugh, 1968; Schiller, 1952; Yerkes, 1943).
In the research reported here, we probe the origins of
snake fear in humans by examining infants’ response to
snakes and other unfamiliar animals. The goal was to
provide evidence relevant to evaluating two influential
accounts of the prevalence of snake fear in humans.
Both theories propose that snake fear originated in early
mammalian evolution when reptiles constituted a wide-
spread, recurrent threat to survival. Fear of snakes would
have led to avoidance of these potentially dangerous
animals, thereby reducing mortality risks and enhancing
the likelihood of survival and reproduction. Where the
two accounts differ is with respect to whether humans have
an
innate fear
of snakes (Menzies & Clarke, 1995; Poulton
& Menzies, 2002) or an
evolved tendency to associate
snakes with fear
(Marks, 1987; Ohman & Mineka, 2001,
2003; Seligman, 1970).
According to what we refer to as the
non-associative
view
(Menzies & Clarke, 1995; Poulton & Menzies,
2002), snake fear is one of several innate, universal fears,
including fear of heights, water, spiders, and strangers.
Learning is not required for the emergence of these
fears: Rather, ‘most members of a species will show fear
to a set of biologically relevant stimuli from early
encounters . . . without any relevant associative learning
experiences’ (Poulton & Menzies, 2002, pp. 127–128).
Consistent with this view, individuals with snake fear
or phobia often cannot identify any learning experiences
that might account for the origin of their fear (Menzies
& Clarke, 1995; Poulton & Menzies, 2002). The explanation
offered for the fact that snake fear (and other purportedly
innate fears) is not present in everyone is that people
‘learn to
not fear
’. Through non-traumatic encounters
with live snakes or representations of snakes, snake fear
is gradually extinguished (Menzies & Clarke, 1995;
Poulton & Menzies, 2002; Rachman & Seligman, 1976;
Rachman, 2002). ‘The . . . role of the environment is to
abate biologically relevant fears, rather than account for
their emergence’ (Menzies & Clarke, 2005, p. 128).
According to prepared learning theory, originally
formulated by Seligman (1970), and what we will refer
to as the
associative-bias account
of Ohman and Mineka
(2001, 2003), the prevalence of snake fear reflects an
innate predisposition to associate snakes with fear. One
form of empirical support for this perspective comes
from Pavlovian fear-conditioning studies with adult
humans (see Ohman & Mineka, 2001, for a review). For
one thing, conditioned skin conductance responses
(SCR) – a measure of emotional activation – are more
resistant to extinction when evolutionarily fear-relevant
stimuli, including snakes and spiders, serve as the CS
Address for correspondence: Judy S. DeLoache, University of Virginia, P.O. Box 400400, Charlottesville, VA 22904, USA; e-mail: jdeloache@
virginia.edu
202 Judy S. DeLoache and Vanessa LoBue
© 2008 The Authors. Journal compilation © 2008 Blackwell Publishing Ltd.
than when the CS is not fear-relevant (e.g. flowers and
mushrooms). For another, fear-relevant responses (e.g.
skin conductance, heart rate acceleration) are more readily
conditioned to fear-relevant stimuli, including snakes,
than to non-fear-relevant stimuli.
Further, there is evidence for
non-conscious
SCR
conditioning to snake stimuli (Ohman & Soares, 1998).
When pictures of snakes and non-threat stimuli were
presented very briefly and followed immediately by a
masking stimulus (making participants unaware of seeing
the pictures), conditioning occurred to the snakes, but
not to the non-threat stimuli. According to Ohman and
Mineka (2003), these findings are consistent with the idea
that ‘responses to snakes are organized by a specifically
evolved primitive neural circuit’ (p. 7).
There is also evidence of enhanced
detection
of snakes.
Adults find a single snake picture among a set of neutral
distracters (e.g. flowers) more rapidly than a single
neutral stimulus among an array of snakes (Ohman, Flykt
& Esteves, 2001).
The strongest evidence in favor of the associative-bias
account of snake fear comes from research with non-human
primates (Cook & Mineka, 1987, 1989). Monkeys that
grow up in the wild are afraid of snakes, but laboratory-
reared monkeys are not, suggesting that wild monkeys
acquired their fear.
In a test of this learning account, lab-reared rhesus
monkeys (adults and adolescents) observed a wild-reared
monkey react with fear to a snake. Simply from observing
a con-specific’s fearful behavior, the monkeys very rapidly
developed an intense, long-lasting fear of snakes them-
selves (Cook & Mineka, 1987). Even viewing a video of
another monkey reacting fearfully to a snake instilled a
fear of snakes (Cook & Mineka, 1989). A crucial finding
in this research is that the vicarious acquisition of fear is
selective
: When shown a video of a monkey reacting
fearfully to a rabbit, a new group of monkeys did not
acquire a fear of rabbits (Cook & Mineka, 1989).
In the research reported here, we introduce empirical
evidence with infants relevant to the debate on the origins
and nature of human snake fear. In the first experiment,
we simply ask whether infants respond differently to
films of snakes versus other animals. In the second, we
ask whether infants show any tendency to associate snakes
with a fearful stimulus. Specifically, do infants associate
the sight of a snake with the sound of a frightened human
voice? (Ethical considerations precluded replicating with
human infants the observational fear-learning studies
done with monkeys.) The third study is a replication
of the second, except that still photographs are substituted
for the snake and animal films used in the second.
Three outcomes were possible in these studies. If
humans do have an
innate fear of snakes
, infants might
display some kind of spontaneous negative reaction to
them. Instead, if there is no innate fear but an
associative
bias
, there should be no initial fearfulness of snakes, but
a tendency to associate snakes with fear-relevant stimuli
should be observed. Further, if humans have a bias to
detect the presence of snakes, differences in orienting to
snakes versus other animals should occur. Finally, if
snakes have no special status, contrary to both accounts
described above, then infants should not respond differ-
entially to snakes compared to other animals.
Experiment 1
To see if human infants would react differently to snakes
versus other animals, we presented 9- to 10-month-olds
with silent films of snakes and exotic non-snake animals.
We used films rather than photographs on the assumption
that snakes’ unique pattern of motion might be critical
in humans’ reactions to them.
One dependent measure was looking time to the two
types of stimuli. If snake fear is innate in humans,
infants might look differentially at snakes and other
animals. For example, they might avoid looking at
snakes and look longer at the non-snakes, or they might
show hyper-vigilance toward snakes, looking longer at
them than at other animals. The second measure was the
infants’ manual exploration of the images on the video
screen. Previous research has shown that 9-month-olds
occasionally feel, pat, rub, and even attempt to grasp at
kinetic video images (Pierroutsakos & Troseth, 2003). It
seemed reasonable to expect that this measure might be
particularly sensitive to any tendency to react differentially
to snakes versus other animals. If infants are in fact
afraid of snakes, they might be reluctant to reach towards
and grasp the image of a moving snake on a television
screen.
Method
Participants
The 16 participants were 9- to 10-month-old infants,
nine males and seven females (
M
= 9.8 mos,
r
= 9.3–
10.5 mos). This age group was selected because of the prior
research demonstrating manual exploration of video
images by 9-month-olds (Pierroutsakos & Troseth, 2003).
The infants were randomly assigned to two stimulus
orders. The participants in this and the second experiment
reported here were recruited from birth announcements
in the local community and were from predominantly
middle-class Caucasian families. Three additional par-
ticipants were excluded from the study (two for fussiness,
and one for experimenter error).
Stimuli
The stimuli were 10 8-sec color film clips from nature
programs in which a snake slithered or an animal walked
at approximately the same slow rate across the screen.
Each of the four snake films showed a different snake.
The six animal films were of wild animals (giraffe,
rhinoceros, polar bear, hippopotamus, elephant, and
Association of snakes and fear 203
© 2008 The Authors. Journal compilation © 2008 Blackwell Publishing Ltd.
large bird
1
). (Although some of these are dangerous
animals, only snakes and spiders are considered to be
evolutionarily relevant threat stimuli.)
Apparatus
The films were presented on a 50.8 cm monitor that was
25 cm from the infant’s eyes. One video camera behind
and a second to the left of the participant recorded looking
at the films and manual exploration of the images.
Procedure
The session started with a warm-up trial of a toy moving
across the screen. At the beginning of each of the 10
trials, a 3-sec attention-getter attracted the infant’s
attention to the screen. The snake and animal clips were
presented in two pseudo-random orders, one the reverse
of the other, with no two snake videos appearing in suc-
cession. Twice during each clip, a female voice prompted
the infant to look at the screen (i.e. ‘Look at that!’).
Results
The results were straightforward. One-way Mixed-Effects
ANOVAs revealed no significant differences in the infants’
behavior toward the two types of stimuli on either of the
two measures (Looking time: Animals – 5.74 sec, Snakes
– 5.62 sec,
F
(1, 15) = 0.05,
ns
; Manual Exploration:
Animals – 0.76, Snakes – 0.61,
F
(1, 15) = 0.18,
ns
).
(Preliminary analyses had revealed no differences for age
group or gender, so these variables were not included in
the analysis.) Power analyses indicated 95% percent
power for both measures.
Discussion
The infants in Experiment 1 did not react differentially
to the snakes and other animals either in terms of how
long they looked at them or how frequently they manu-
ally explored them. The results thus provide no support
for the non-associative account, that is, no evidence of
an innate fear of snakes. Any differential responding of
any sort could have offered support for this view: The
infants might have looked longer at the snakes, indicating
this category of stimulus was prepotent, or they might have
avoided looking at snakes, suggesting that these stimuli
were aversive. However, it is important to point out that
the absence of differential responding is not definitive: It
has been suggested that an innate fear of snakes might
be manifested only after infants have begun walking and
moving around in the environment (Marks, 1987). The
results of Experiment 2 speak to this possibility.
The lack of differential responding in Experiment 1 is
consistent with the associative-bias account. According
to this view, no negative reactions would be expected to
the snake images in the absence of any fearful stimuli
with which to associate them. This null result does not,
however, provide strong support for this account.
Experiment 2
Informed by the results of Experiment 1, the second study
was designed to provide a
direct
test of the associative-
bias hypothesis. If humans have an innate predisposition
to learn to fear snakes, that predisposition should exist
independent of experience with snakes. Hence, a
tendency
to associate a fearful stimulus with snakes
might be observ-
able even in infants. We reasoned that such a tendency
might lead infants to pay more attention to snakes when
they heard a frightened human voice than when they
heard a happy voice.
To examine this possibility, we used an auditory-visual
matching paradigm. This well-established procedure is
based on the often-replicated tendency of infants (and
adults) to look selectively at a visual display that matches
an auditory stimulus (e.g. Golinkoff, Hirsh-Pasek, Cauley
& Gordon, 1987; Spelke & Cortelyou, 1981; Walker-
Andrews, Bahrick, Raglioni & Diaz, 1991). Of particular
relevance here, infants match the
emotional valence
of
voices and visual stimuli. For example, infants as young
as 5 months of age who are presented with pairs of faces
displaying different emotional expressions look longer
at a smiling face when they hear a happy voice and
longer at an angry face when listening to an angry voice
(Walker-Andrews, 1986).
Further, infants use another person’s vocal emotion
as a source of information for evaluating situations
(Mumme, Fernald & Herrera, 1986; Sorce, Emde, Campos
& Klinnert, 1985; Tamis-LeMonda, Adolph, Dimitro-
poulou & Zack, 2006). They rely on emotional tone even
when a message is delivered in an unfamiliar language
(Fernald, 1992).
As Figure 1 depicts, the infants in Experiment 2 were
simultaneously presented with two films – one of a snake
and the other of an exotic animal – accompanied by a
recording of either a very frightened or a very happy human
voice speaking in a nonsense language (Banse & Scherer,
1996). If infants have a predisposition to associate snakes
with fear, as proposed by prepared-learning theory
(e.g. Ohman & Mineka, 2001; Seligman, 1970), they should
look longer at snakes when they hear a frightened voice
than when they hear a happy voice. Note: we did not
expect the infants to react fearfully to the snakes,
only to look more at them in the presence of a fearful
voice.
There is no basis for making any prediction regarding
how long the infants would look at the non-snake animals
1
Fewer snake than animal films were used out of concern that the
infants might habituate more rapidly to the snakes than to the animals,
which differed more in terms of appearance and pattern of movement.
A comparison of the eight infants who had received equal numbers
of snakes and non-snake animals in their first six trials revealed no
differences in either dependent variable for the two types of stimuli.
204 Judy S. DeLoache and Vanessa LoBue
© 2008 The Authors. Journal compilation © 2008 Blackwell Publishing Ltd.
as a function of emotional voice. Hence, the predicted
result is an interaction of type of animal by vocal emotion,
with significantly longer looking to snakes while hearing
a frightened voice than a happy one.
This design also enabled us to examine whether, like
adults in threat-detection tasks, infants more readily
detect the presence of snakes than other animals. If
snakes represent a threat-relevant category for infants, a
bias should be apparent in terms of infants’ orienting
behavior to the snakes versus the other animals.
Method
Participants
Participants were 48 infants, half 7- to 9-month-olds
(
M
= 8.0 mos,
r
= 7.2–8.8 mos) and half 16- to 18-
month-olds (
M
= 16.4 mos,
r
= 14.8–18.3 mos). All but
one of the infants in the older group were reported to be
walking. An additional 15 infants failed to complete the
experiment, most because of fussiness. Half the males
and half the females at each age were randomly assigned
to one of two stimulus orders.
Stimuli
Films.
Twelve films (six snakes and six exotic animals)
were used (including the 10 from Experiment 1 plus two
additional snake clips).
Voices.
The 12 professionally produced audio record-
ings were of the same two nonsense phrases (‘Hat sundig
pron you venzy. Fee gott laish jonkill gosterr’) spoken by
two men and two women. Most importantly, one recording
made by each person was in a pleasant, happy-sounding
tone of voice, and the other sounded distinctly frightened.
(These recordings have been scaled for emotional con-
tent and used in many studies of adult perception of
emotion; Banse & Scherer, 1996.)
Film–voice pairings.
Twelve film–voice combinations
were constituted by (a) randomly selecting one snake
and one animal film to form each of six film pairs and
(b) randomly assigning each film pair to one of the
frightened voices and one of the happy voices. Each
film–voice pair was converted to a single QuickTime
video, with the two clips side by side. The snakes and
non-snake animals appeared equally often on the right
and left side. Both animals began moving simultaneously
and at approximately the same speed. The voices came
on 3 sec before the onset of the films and continued
throughout the 10-sec films.
Stimulus presentation.
As Figure 1 shows, the films
were projected onto a 91.4 cm by 121.9 cm white screen
approximately 91 cm from the infant. The projected
films measured 48.3
×
30.5 cm with 30.4 cm between
them. The voices came from two speakers located on
either side of the screen. A video camera filmed the
infant’s head and eyes through a small hole in the screen.
Each infant received 12 trials, each involving a different
one of the 12 film–voice combinations.
The experimenter observed the infant’s looking
behavior on a monitor in the adjoining room in order to
control the beginning of each trial. In between trials, the
infant’s attention was attracted by a blinking green dot
appearing in the center of the screen accompanied by a
‘dinging’ sound. The blinking light automatically
appeared at the end of each film, but the experimenter
initiated the onset of the next trial on the computer
controlling the presentation.
Procedure
Each infant was seated on a parent’s lap in front of the
screen. The parent was blindfolded to preclude any in-
advertent cues. The experimenter manually began each trial
as soon as the attention-getter attracted the infant’s gaze to
the screen. Everything else was controlled automatically.
Coding
The Supercoder program (Hollich, Rocroi, Hirsh-Pasek
& Golinkoff, 1999) was used to code looking time for
each infant. The coding of the digital tapes was done
frame by frame at a rate of 30 frames per second, making
the coding very sensitive. All coding was blind: Because
the videos were silent and showed only the infants’
Figure 1 The infant sat on the blindfolded parent’s lap. On
each trial, a pair of films of a snake and another animal was
presented, accompanied by a tape of either a frightened or
happy voice.
Association of snakes and fear 205
© 2008 The Authors. Journal compilation © 2008 Blackwell Publishing Ltd.
faces, the coders could not know which voice the infant
was hearing or on which side of the screen the snakes
and animals appeared. The coders recorded the latency
from the onset of the films to the infant’s first fixation of
one of them, and the total amount of time the infant
looked at each of the two visual stimuli. All of the tapes
were coded by a primary coder, and one-fourth of the
tapes were independently coded by a second coder.
Agreement between coders was 80% for latency and 81%
for looking time (an acceptable level of reliability in
infant attention research). When coders disagreed, the
primary coder’s codes were used.
2
Results
The looking-time data are shown in Figure 2. They
reveal the predicted result – significantly longer looking
at snakes while hearing a fearful voice than a happy one.
As noted before, there is no basis for predicting differential
looking at the non-snake animals as a function of vocal
emotion, and no difference occurred.
The primary data analysis was a 2 (snake vs. animal)
×
2 (frightened vs. happy voice) Mixed Effects analysis
of variance (Bagiella, Sloan & Heitjan, 2000; Gueor-
guieva & Krystal, 2004). (Preliminary analyses had
revealed no differences for age or gender, so these variables
were not included in the analysis.) The only significant
result was the predicted interaction between type of
animal and type of vocal emotion,
F
(1, 1077) = 4.56,
p
< .04. Post-hoc comparison revealed that, as predicted,
the infants looked longer at the snakes when listening
to a frightened voice than when listening to a happy
one,
F
(1, 515) = 3.98,
p
< .05. In contrast, the post-hoc
comparison of looking times to the other animals as a
function of vocal emotion was not significant. These
results are consistent with the hypothesis that infants
would associate the sound of a frightened voice with the
sight of a snake.
Discussion
The overall pattern of results in Experiment 2 revealed
important differences in infants’ responses to snakes.
Primarily, the looking-time data revealed a significant
tendency to associate snakes with fearful stimuli: The
infants looked longer at the snakes when listening to a
frightened voice than a happy one. This result is consistent
with the associative-bias view. Infants also turned
somewhat more rapidly to look at the snakes than at the
non-snakes.
Another aspect of the results of this study worth noting
is the
absence
of a significant age difference in the two
groups of infants (8- and 16-month-olds). This result
indicates that the tendency to associate snakes with fear
emerges well before the onset of walking. More impor-
tantly, the highly similar pattern of behavior for infants
only slightly more than half a year old and those a full
year older supports the idea of a predisposition that is
independent of both maturational status and experience.
Experiment 3
The results of Experiment 2 raise an important question
concerning what stimulus attributes of snakes underlie
infants’ tendency to associate them with a fearful stimulus.
One likely candidate is their sinusoidal movement
pattern; no other terrestrial animals move like snakes do.
If the unique pattern of movement associated with snakes
is a crucial determinant of humans’ negative reactions to
them, one would predict that infants would not associate
stationary
images of snakes with fearful voices. Accord-
ingly, in Experiment 3, we replicated Experiment 2,
except that the visual stimuli were still photographs of
snakes and other exotic animals. The voices were the
same as those used in Experiment 2.
Method
Participants
Participants were 48 infants, half 7- to 9-month-olds (
M
= 9.2 mos,
r
= 7.3–11.7 mos) and half 16- to 18-month-
olds (
M
= 17.2 mos,
r
= 16.1–18.6 mos). An additional
13 infants failed to complete the experiment, most
because of fussiness. Half the males and half the females
2
Any data points that were 3 times the interquartile range outside of
the interquartile range and were more than three standard deviations
above or below the mean (Cohen, Cohen, West & Aiken, 2003) were
identified as outliers and eliminated from the data. A total of 8 and 10
outliers were identified for Experiments 2 and 3, respectively. The elimin-
ated data points constituted 1.6% of the data.
Figure 2 As predicted, in Experiment 2, the infants looked
significantly longer at the snakes when listening to a frightened
-
sounding voice than when listening to a happy voice. Looking
times to the other animals did not differ significantly for the
happy and frightened voices. In Experiment 3, looking times
did not differ for either the snakes or the other animals for the
happy and frightened voices. Error bars = SE of the mean.
206 Judy S. DeLoache and Vanessa LoBue
© 2008 The Authors. Journal compilation © 2008 Blackwell Publishing Ltd.
at each age were randomly assigned to one of two stimulus
orders.
Stimuli
The stimuli used in Experiment 3 were 12 still photo-
graphs of snakes and exotic animals taken from nature
books. The projected images were the same size as the
videos used in Experiment 2. The animal photographs
matched the films in Experiment 2 in that they were of
the same types of animals of the same colors and against
generally the same types of background. The voices were
identical to those of Experiment 2, and the stimulus
pairings were constituted in the same way.
Procedure
The procedure for Experiment 3 was identical to that of
Experiment 2. Coding was the same as that of Experiment
2, with agreement between coders of 80% for latency and
81% for looking time.
Results
The looking-time results are shown in Figure 2. As
expected and in contrast to the interaction that was
present in Experiment 2, the looking behavior of these
infants did not differ for snakes versus other animals
based on what voice they were hearing.
The looking-time data were analyzed in a 2 (snake vs.
non-snake animal)
×
2 (frightened vs. happy voice) Mixed
Effects analysis of variance. (Preliminary analyses had
revealed no differences for age or gender, so these variables
were not included in the analysis.) The only significant
result was a main effect of animal,
F
(1, 1098) = 66.34,
p
< .01. Consistent with the results of Experiment 2, the
infants looked longer at the non-snakes than the snakes.
There was no interaction of animal by voice,
F
(1, 1098)
= 0.11,
ns
. The infants did not look differentially at
the non-moving snakes and other animals. Of most
importance, they did not look longer at a still snake than
at a still animal when they heard a frightened voice.
Additionally, examination of the infants’ first looks
revealed that they turned significantly more rapidly to the
snakes (
M
= 0.74 s) than to the non-snakes (
M
= 0.91 s),
F
(1, 511) = 3.82,
p
= .05. There were no other significant
effects. (The same pattern of behavior also occurred in
Experiment 2, but the difference was not significant.)
General discussion
The three studies reported here provide evidence sup-
porting the prepared-learning/associative-bias account
of the prevalence of snake fear and phobia, specifically,
the claim that humans are predisposed to learn to asso-
ciate snakes with fear. The current results offer particu-
larly strong support by virtue of the fact that such a
predisposition was observed in infants. In the main
study, Experiment 2, the prediction that infants in the
first and second years of life would associate the sight of
a moving snake with the sound of a fearful voice was
confirmed.
In contrast to this predicted positive result, infants in
Experiment 1 did not respond differentially when films
of snakes and other unfamiliar animals were presented
individually, unaccompanied by voices. Further, in
Experiment 3, in which pairs of still photographs of snakes
and other animals were accompanied by emotional
voices, infants did not respond differentially. Thus, the
overall pattern of the results reveals a natural tendency
in human infants to associate the sight of an undulating
snake with the sound of a frightened human voice.
The results of Experiments 2 and 3 also provided
evidence of an attentional bias in infants’ response to
snakes versus other unfamiliar animals. Consistent with
research with adults (Ohman
et al.
, 2001), the snakes
more readily recruited the infants’ attention than the
other animals did, as shown by more rapid orienting to
snakes when the stimulus pairs first appeared. This
preferential orienting to snakes constitutes the first
evidence that, like adults (Ohman
et al.
, 2001) and older
children (LoBue & DeLoache, 2008), infants respond
to the presence of a snake more rapidly than other kinds
of animals. The more rapid turning to snakes suggests
that when a snake is detected, infants are biased to
respond to it very quickly.
The present studies raise several questions for future
research. One is whether the same results as in the experi-
ments reported here would be found with spiders –
another fear-relevant class of stimuli. Like snakes, spiders
are a very frequent target of fears and phobias, and there
is substantial evidence with adults for conditioning and
detection effects with spiders similar to those reported
for snakes (Ohman & Mineka, 2001, 2003). Thus, infants
might show a tendency to associate spiders with fearful
voices.
Another important issue for further research concerns
the role of the unique movement pattern of snakes in
eliciting negative responses. In Experiment 3, we established
that still photographs of snakes and other animals do
not elicit differential responding, suggesting that move-
ment may play an important role in humans’ reactions
to snakes. We are currently examining the importance of
motion pattern
on its own
by replicating Experiment 2
using point light displays of snakes and other animals.
In summary, the research reported here provides the
first evidence with infants relevant to the origins of one
of the most common fears and phobias present in humans.
The results offer support for the view that humans have
a predisposition to associate snakes with fear. The
results also indicate that the unique, anomalous movement
pattern of snakes may underlie this association.
These findings suggest that it was the slithering
motion of the ‘narrow fellow in the grass’ that aroused
Emily Dickinson’s fear.
Association of snakes and fear 207
© 2008 The Authors. Journal compilation © 2008 Blackwell Publishing Ltd.
Acknowledgements
We gratefully thank Themba Carr, Christina Danko,
Daniel Draschler, Lindsey Doswell, Kai Van Eron, and
Greg Clumpner for assistance with the research, Jack
McArdle for statistical advice, and George Hollich for
invaluable assistance regarding procedures and coding
for Experiments 2 and 3. We thank Klaus Scherer for pro-
viding the voice stimulus set (which is based on research by
Klaus Scherer, Harald Wallbott, Rainer Banse, and Heiner
Ellgring; see Banse & Scherer, 1996). Beatrice van Gelden
also provided helpful assistance with stimuli. Colleagues
Gerald Clore, Bobbie Spellman, and Bethany Teachman
provided helpful comments on the manuscript. This research
was partially supported by NIH grant HD-25271.
References
Agras, S., Sylvester, D., & Oliveau, D. (1969). The epidemiology
of common fears and phobias.
Comprehensive Psychiatry
,
10
,
151–156.
Bagiella, E., Sloan, R.P., & Heitjan, D.F. (2000). Mixed-effects
models in psychophysiology.
Psychophysiology
,
37
, 13–20.
Banse, R., & Scherer, K.R. (1996). Acoustic profiles in vocal
emotion expression.
Journal of Personality and Social Psy-
chology
,
70
, 614–636.
Cohen, J., Cohen, P., West, S.G., & Aiken, L.S. (2003).
Applied
multiple regression/ correlation analysis for the behavioral sciences
(3rd edn.). New York: Erlbaum.
Cook, M., & Mineka, S. (1987). Second-order conditioning
and overshadowing in the observational conditioning of
fear in monkeys.
Behaviour Research and Therapy
,
25
, 349–
364.
Cook, M., & Mineka, S. (1989). Observational conditioning of
fear to fear-relevant versus fear-irrelevant stimuli in rhesus
monkeys.
Journal of Abnormal Psychology
,
98
, 448–459.
Fernald, A. (1992). Meaningful melodies in mothers’ speech to
infants. In H. Papousek & U. Jürgens (Eds.),
Nonverbal vocal
communication: Comparative and developmental approaches
(pp. 262–282). New York: Cambridge University Press.
Fredrikson, M., Annas, P., Rischer, H., & Wik, G. (1996).
Gender and age differences in the prevalence of specific fears
and phobias.
Behaviour Research and Therapy
,
34
, 33–39.
Golinkoff, R.M., Hirsh-Pasek, K., Cauley, K.M., & Gordon, L.
(1987). The eyes have it: lexical and syntactic comprehension
in a new paradigm.
Journal of Child Language
,
14
, 23–45.
Gueorguieva, R., & Krystal, J. (2004). Move over ANOVA:
progress in analyzing repeated-measures data and its
reflection in papers published in the Archives of General
Psychiatry.
Archives of General Psychiatry
,
61
, 310–317.
Hollich, G., Rocroi, C., Hirsh-Pasek, K., & Golinkoff, R.M.
(1999). Testing language comprehension in infants: introducing
the split-screen preferential looking paradigm. Poster pre-
sented at SRCD, Albuquerque, NM, April.
Joslin, J., Fletcher, H., & Emlin, J. (1964). A comparison of the
responses to snakes of lab- and wild-reared rhesus monkeys.
Animal Behavior
,
12
, 348–352.
King, G.E. (1997). The attentional basis for primate responses
to snakes. Paper presented at the annual meeting of the
American Society of Primatologists, San Diego, CA.
LoBue, V., & DeLoache, J.S. (2008). Detecting the snake in the
grass: attention to fear-relevant stimuli by adults and young
children.
Psychological Science
,
19
, 284–289.
Marks, I. (1987).
Fears, phobias, and rituals: Panic, anxiety, and
their disorders
. Oxford: Oxford University Press.
Menzies, R.G., & Clarke, J.C. (1995). The etiology of phobias: a
nonassociative account.
Clinical Psychology Review
,
15
, 23–48.
Mumme, D.L., Fernald, A., & Herrera, C. (1996). Infants’
responses to facial and vocal emotional signals in a social
referencing paradigm.
Child Development
,
67
, 3219–3237.
Ohman, A., Flykt, A., & Esteves, F. (2001). Emotion drives
attention: detecting the snake in the grass. Journal of Experi-
mental Psychology: General, 130, 466–478.
Ohman, A., & Mineka, S. (2001). Fears, phobias, and pre-
paredness: toward an evolved module of fear and fear
learning. Psychological Review, 108, 483–522.
Ohman, A., & Mineka, S. (2003). The malicious serpent:
snakes as a prototypical stimulus for an evolved module of fear.
Current Directions in Psychological Science, 12, 5–8.
Ohman, A., & Soares, J.J.F. (1993). On the automatic nature
of phobic fear: conditioned electrodermal responses to masked
fear-relevant stimuli. Journal of Abnormal Psychology, 102,
121–132.
Pierroutsakos, S.L., & Troseth, G.L. (2003). Video verite:
infants’ manual investigation of objects on video. Infant
Behavior and Development, 26, 183–199.
Poulton, R., & Menzies, R.G. (2002). Non-associative fear-
acquisition: a review of the evidence from retrospective and
longitudinal research. Behaviour Research and Therapy, 40,
127–149.
Rachman, S. (2002). Fears born and bred: non-associative fear
acquisition. Behaviour Research and Therapy, 40, 121–126.
Rachman, S., & Seligman, M.E.P. (1976). Unprepared phobias:
be prepared. Behaviour Research and Therapy, 14, 333 –338.
Rumbaugh, D.M. (1968). The learning and sensory capacities
of the squirrel monkey in phylogenetic perspective. In
L.A. Rosenblum & R.W. Cooper (Eds.), The squirrel monkey
(pp. 256–317). New York: Academic Press.
Schiller, P. (1952). Innate constituents of complex responses.
Psychological Review, 59, 177–191.
Seligman, M. (1970). On the generality of laws of learning.
Psychological Review, 77, 406–418.
Sorce, J.F., Emde, R.N., Campos, J., & Klinnert, M.D. (1985).
Maternal emotional signaling: its effect on the visual cliff
behavior of 1-year-olds. Developmental Psychology, 21, 195–200.
Spelke, E.S., & Cortelyou, A. (1981). Perceptual aspects of social
knowing: looking and listening in infancy. In M.E. Lamb
& L.R. Sherrod (Eds.), Infant social cognition (pp. 61–84).
Hillsdale, NJ: Erlbaum.
Tamis-LeMonda, C.S., Adolph, K.E., Dimitropoulou, K.A., &
Zack, E.A. (2007). Locomotor experience and use of social
information are posture specific. Unpublished manuscript.
Walker-Andrews, A.S. (1986). Intermodal perception of
expression behaviors: relation of eye and voice? Developmental
Psychology, 22, 373–377.
Walker-Andrews, A.S., Bahrick, L.E., Raglioni, S.S., & Diaz, I.
(1991). Infants’ bimodal perception of gender. Ecological
Psychology, 3, 55–75.
Yerkes, Robert M. (1943). Chimpanzees: A laboratory colony.
New Haven, CT: Yale University Press.
Received: 15 January 2007
Accepted: 26 January 2008
