INFORME
Is Empiricism Innate? Preference for Nurture Over
Nature in People’s Beliefs About the Origins
of Human Knowledge
Jinjing (Jenny) Wang
1 and Lisa Feigenson1
1Department of Psychological and Brain Sciences, Universidad Johns Hopkins
un acceso abierto
diario
Palabras clave: nature-nurture, intuitive theories, core knowledge, nativism, empiricism
ABSTRACTO
The origins of human knowledge are an enduring puzzle: what parts of what we know
require learning, and what depends on intrinsic structure? Although the nature-nurture
debate has been a central question for millennia and has inspired much contemporary
research in psychology and neuroscience, it remains unknown whether people share
intuitive, prescientific theories about the answer. Here we report that people (norte = 1, 188)
explain fundamental perceptual and cognitive abilities by appeal to learning and instruction,
rather than genes or innateness, even for abilities documented in the first days of life. A NOSOTROS.
adultos, adults from a culture with a belief in reincarnation, niños, and professional
scientists—including psychologists and neuroscientists, all believed these basic abilities
emerge significantly later than they actually do, and ascribed them to nurture over nature.
These findings implicate a widespread intuitive empiricist theory about the human mind,
present from early in life.
INTRODUCCIÓN
Some questions about human nature persistently captivate our thinking. Among these is the
puzzle of where knowledge comes from: what must be learned from our actions and obser-
vaciones, and what emerges from our biological inheritance? This mystery of how nature and
nurture give rise to the mind has animated debate for millennia (Cooper & hutchinson, 1997),
inspiring research in linguistics, neurociencia, and psychology. Although there is consensus
that some mental abilities are built from experience and instruction, like reading and calcu-
lating square roots, there is disagreement over what innate capacities exist in support of this
aprendiendo. Arguments that some capacities depend on innate structure point to findings of com-
petence in very young infants or individuals lacking relevant learning opportunities (Izard,
Sann, Spelke, & Streri, 2009; Senghas, Kita, & Özyürek, 2004), and to the observation that
even with such opportunities, the available data underspecify what humans seem to learn
(Chomsky, 1980). These arguments face counterclaims that this knowledge instead is acquired
via powerful learning mechanisms, sometimes soon after birth (Skinner, 2014; Verguts & Fias,
2004). Although the nature-nurture debate continues, the empiricist position that knowledge
requires learning has dominated much of the last century (see Margolis, samuel, & Stich,
2012), with some arguing that empiricism should be the default position absent undeniable
evidence otherwise (Príncipe, 2012).
Citación: Wang, J., & Feigenson, l.
(2019). Is Empiricism Innate?
Preference for Nurture Over Nature in
People’s Beliefs About the Origins of
Human Knowledge. Mente abierta:
Descubrimientos en ciencia cognitiva, 3,
89–100. https://doi.org/10.1162/
opmi_a_00028
DOI:
https://doi.org/10.1162/opmi_a_00028
Materiales suplementarios:
https://doi.org/10.1162/opmi_a_00028
Recibió: 24 Febrero 2019
Aceptado: 4 Julio 2019
Conflicto de intereses: Los autores
declare that they have no competing
interests.
Autor correspondiente:
Jinjing (Jenny) Wang
jinjing.jenny.wang@gmail.com
Derechos de autor: © 2019
Instituto de Tecnología de Massachusetts
Publicado bajo Creative Commons
Atribución 4.0 Internacional
(CC POR 4.0) licencia
La prensa del MIT
yo
D
oh
w
norte
oh
a
d
mi
d
F
r
oh
metro
h
t
t
pag
:
/
/
d
i
r
mi
C
t
.
metro
i
t
.
/
mi
d
tu
oh
pag
metro
i
/
yo
a
r
t
i
C
mi
–
pag
d
F
/
d
oh
i
/
i
/
.
1
0
1
1
6
2
oh
pag
metro
_
a
_
0
0
0
2
8
1
9
2
8
7
8
8
oh
pag
metro
/
/
i
–
0
3
–
8
9
pag
d
.
F
b
y
gramo
tu
mi
s
t
t
oh
norte
0
7
S
mi
pag
mi
metro
b
mi
r
2
0
2
3
Is Empiricism Innate? Wang, Feigenson
In this ongoing debate between nativism and empiricism, scientists use the mind to study
the mind. But what if the primary instrument in this endeavor, the human mind, intuitively
favors one explanation of knowledge over the other? Children and adults hold a variety of
intuitive theories that are sometimes at odds with (and may later be replaced by) theories ac-
quired through formal instruction (carey, 2009). These intuitive theories are potent enough
that they sometimes affect scientific theorizing by professionals. An example comes from the
science of object physics. For over a thousand years, from Philoponus in the 6th century to
Galileo in the 16th, philosopher-scientists mistakenly believed that moving objects acquire an
energy that propels them until the energy runs out (McCloskey, 1983). Their “impetus theory,"
incompatible with Newtonian mechanics and contemporary physics, is not just an archaic
vista. Modern adults also make wrong predictions about object motion; strikingly, their pre-
dictions echo those of the premedieval and medieval scientists, even though they were never
taught the impetus theory and had even taken college physics (McCloskey, 1983). This shows
that beliefs about the world, even formal scientific beliefs held by specialists, can reflect folk
intuitions that are broadly shared and resistant to revision. Teniendo esto en cuenta, consider again
the case of where basic human knowledge originates. For this ancient question that continues
to drive current research, generating answers from the dichotomous (“nature or nurture”) a
the more nuanced (“nature and nurture”), do people think of nature and nurture as natural
alternatives? En ese caso, do they intuitively prefer one as the better explanation?
To characterize folk beliefs about nature and nurture, we probed intuitions about aspects
of the mind that have been studied experimentally, thereby providing a “ground truth” against
which to compare. The basic, nonverbal, “core” knowledge that guides everyday actions—
like believing that hidden objects still exist, or that 10 apples are more than five, offers such a
caso, because research of the past 40 years has carefully examined its developmental origins.
Laboratory experiments have illuminated ways in which infants represent and reason about
objects, quantities, and other people by measuring their looking times and neural responses to
these stimuli (Spelke & Kinzler, 2007). Comparing what science has discovered about the ori-
gins of this foundational core knowledge to people’s intuitive theories can reveal both hidden
systematicities and limitations in thinking about thinking.
EXPERIMENT 1
Participantes
One hundred and one adults (m = 34.95 years old; 59 femenino) were recruited through Amazon
Mechanical Turk.
Diseño, Estímulos, and Coding
Participants read a description of a character Alex, with four of her behaviors chosen to illustrate
an innate ability, an ability emerging through biological maturation, an ability learned through
observation and practice, and an explicitly taught ability: (1) Alex was born able to drink milk,
(2) Alex grew older and became strong enough to lift heavy things, (3) Alex learned on her own
how to play on a computer, (4) Alex learned from someone else how to tie her shoes. Entonces
participants were presented seven core perceptual and cognitive abilities, each the subject of
much previous developmental research: color perception (distinguishing two colors), profundidad
percepción (distinguishing a nearby from a distant object), face recognition (distinguishing face-
like from non-facelike things), physical reasoning (thinking an unsupported object will fall), transmisión exterior-
ject permanence (thinking a hidden object still exists), approximate numerical discrimination
(thinking an array of 10 items has more than an array of five), and social evaluation (preferring
helping others to not helping) (Baillargeon, 1987, 1995; Bornstein, Kessen, & Weiskopf, 1976;
MENTE ABIERTA: Descubrimientos en ciencia cognitiva
90
yo
D
oh
w
norte
oh
a
d
mi
d
F
r
oh
metro
h
t
t
pag
:
/
/
d
i
r
mi
C
t
.
metro
i
t
.
/
mi
d
tu
oh
pag
metro
i
/
yo
a
r
t
i
C
mi
–
pag
d
F
/
d
oh
i
/
i
.
/
1
0
1
1
6
2
oh
pag
metro
_
a
_
0
0
0
2
8
1
9
2
8
7
8
8
oh
pag
metro
/
/
i
–
0
3
–
8
9
pag
d
.
F
b
y
gramo
tu
mi
s
t
t
oh
norte
0
7
S
mi
pag
mi
metro
b
mi
r
2
0
2
3
Is Empiricism Innate? Wang, Feigenson
yo
D
oh
w
norte
oh
a
d
mi
d
F
r
oh
metro
h
t
t
pag
:
/
/
d
i
r
mi
C
t
.
metro
i
t
.
/
mi
d
tu
oh
pag
metro
i
/
yo
a
r
t
i
C
mi
–
pag
d
F
/
d
oh
i
/
i
/
.
1
0
1
1
6
2
oh
pag
metro
_
a
_
0
0
0
2
8
1
9
2
8
7
8
8
oh
pag
metro
/
/
i
–
0
3
–
8
9
pag
d
.
F
b
y
gramo
tu
mi
s
t
t
oh
norte
0
7
S
mi
pag
mi
metro
b
mi
r
2
0
2
3
Cifra 1. Methods and results for Experiment 1. (A) Example responses to “How come Alex can
[tell colors apart]?” and categories of coded responses. (B) Mean responses in Experiment 1. Circles’
left-right positions represent participants’ age estimates. Circles’ colors represent the distribution
of free responses. Green crosses indicate the earliest age category at which abilities have been
documented in published research.
Farroni et al., 2005; Hamlin, Wynn, & Bloom, 2007; Izard et al., 2009; Slater, Mattock, &
Marrón, 1990) (Table S1, Wang & Feigenson, 2019). Participants indicated when they thought
each ability first was present by clicking an image depicting Alex as a newborn, older infant,
toddler, preschool-age child, school-age child, adolescent, or adult. Participants rarely chose
these last two images; por lo tanto, we excluded them from the figures’ x-axes but included any
responses to them in all analyses. Participants also were asked “how come” Alex had each abil-
idad, and typed a free response. To confirm that participants understood the task, we included
three anchor items involving abilities predicted to generate consensus as being present early in
life and requiring little or no learning (seeing and hearing), or as appearing later and requiring
learning and instruction (lectura).
We quantified participants’ timeline choices by first estimating Alex’s age in each time-
line image using the midpoint between the labeled lower and upper age bounds (Figura 1B).
We fit these ages using their ordinal positions, resulting in the function y = 0.13e0.78x, R2 > .99.
Using this equation, we translated participants’ timeline responses into an average age onset
estimate for each ability, allowing us to compare participants’ estimates to the average onset
age suggested by published findings (Table S2, Wang & Feigenson, 2019).
Participants’ free responses to the question “how come Alex can X” were coded into
four categories (Figura 1A). Explanations that an ability “was innate” or was due to biological
MENTE ABIERTA: Descubrimientos en ciencia cognitiva
91
Is Empiricism Innate? Wang, Feigenson
estructura (p.ej., “she can see because she has eyes,” “it’s in her genes”), or that a person “was just
born able to X” were coded as endorsing innateness. Explanations that an ability “emerged on
its own” or “happens as a person gets older” were coded as endorsing maturation. Explanations
that a person “became able to X through their own observations,” or “learned X on their own,"
were coded as endorsing learning without explicit instruction. Explanations that an ability “was
taught by others” or “was learned in school” were coded as endorsing explicit teaching. De
1,010 respuestas, 15% did not clearly fall into any category (p.ej., “kids have this ability at this
edad,” “she is smart”), and were excluded from analysis.
Resultados
Participants estimated that core knowledge abilities emerged, on average, entre 0.94 y
2.88 years of age (Figura 1B) (95% CIs [0.66, 1.22] [2.60, 3.15]), more than eight times the
average age of onset suggested by published findings. Strikingly, participants overwhelmingly
explained the core abilities using learning-based explanations (explanations endorsing learn-
ing without instruction plus explanations endorsing explicit teaching), at levels exceeding
chance (m = 77%; 95% CI [73%, 81%]; t(99) = 11.81, pag < .001) (Figure 1B). For no core
ability did learning-based responses drop below 50%, even for the more perceptual abilities of
distinguishing colors (62%) and distances (69%; binomial exact tests ps < .05). For the remain-
ing core abilities, learning-based responses were even more prevalent, averaging 82%; 95%
CI [77%, 87%] (see the Supplemental Materials, Wang & Feigenson, 2019, for analyses of the
influence of gender, age, educational background, parental status, and relevant coursework
on participants’ responses). Hence, adults believed that basic perceptual and cognitive abili-
ties, suggested by research to be available in early infancy, are acquired much later through
learning and instruction.
This was not because participants simply pointed to images of older children, or did not
understand the idea of innateness. For seeing and hearing, adults correctly believed that infants do
both (M = 0.32 years; 95% CI [0.05, 0.58]) (Brown & Yamamoto, 1986; Northern & Downs,
2002), and almost never offered learning-based explanations (M = 3%; 95% CI [0%, 5%]). For
reading, adults believed that this ability appears around early school age (M = 4.56 years; 95%
CI [4.29, 4.83]) (Hasbrouck & Tindal, 2011), and 100% gave learning-based explanations.
Two replication studies tested alternative explanations of our findings, confirming that
the observed pattern was not due to participants’ perception of the timeline, nor to the criteria
used to code free responses. Participants in Experiment 1b (N = 100) were probed on the same
test items but, for each ability, typed the age at which they thought each ability first was present
(without a timeline), and indicated the ability’s origins by adjusting a sliding bar in response
to the prompt “Where does Alex’s ability to [e.g., tell colors apart] come from?” The bar’s end-
points were 0 (“entirely from her genes”) and 100 (“entirely from her experience”). Participants
in Experiment 1c (N = 202) were tested with the timeline and free response, using ability de-
scriptions with increased or decreased emphasis on metacognitive or verbal capacities. For
example, instead of “tell colors apart,” we asked “When was the first time/ how come Alex’s
brain [responded differently to different colors]?”). In both replications, participants again over-
estimated the core abilities’ onset, and attributed them to learning (Figures S1 and S2, Wang
& Feigenson, 2019).
Is the intuitive empiricism revealed in Experiment 1 culturally specific? To find out, we tested
adults living in India, where approximately 80% of people identify as Hindu. A major tenet of
Hinduism involves reincarnation, a belief that could increase nativist beliefs because aspects
of the self are thought to precede individual experience.
OPEN MIND: Discoveries in Cognitive Science
92
l
D
o
w
n
o
a
d
e
d
f
r
o
m
h
t
t
p
:
/
/
d
i
r
e
c
t
.
m
i
t
.
/
e
d
u
o
p
m
i
/
l
a
r
t
i
c
e
-
p
d
f
/
d
o
i
/
i
.
/
1
0
1
1
6
2
o
p
m
_
a
_
0
0
0
2
8
1
9
2
8
7
8
8
o
p
m
/
/
i
-
0
3
-
8
9
p
d
.
f
b
y
g
u
e
s
t
t
o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3
Is Empiricism Innate? Wang, Feigenson
EXPERIMENT 2
Participants
Ninety-nine adults living in India, who self-identified as Hindu (M = 29.49 years; 22 female),
were recruited through Amazon Mechanical Turk.
All aspects were as in Experiment 1. In addition, participants
Design, Stimuli, and Coding
reported their religiosity on a 5-point scale (1 = very religious, 5 = not at all religious). Of 990
free responses, 36% did not fall into any category and were excluded.
Results
Like U.S. participants, Indian participants believed core abilities emerge later than they actually
do, choosing onsets between 2.25 and 5.11 years (95% CIs [1.98, 2.53] [4.84, 5.38]), more
than 16 times the age of actual onset, and gave mostly learning-based explanations of the
abilities’ origins (M = 80%; 95% CI [74%, 86%]) (Figure 2). Degree of religiosity predicted
neither onset estimates (β = −0.14, p = .20) nor learning-based explanations (β = 0.032,
p = .77).
Are the intuitively empiricist beliefs of U.S. and Indian adults specific to human abilities,
or is it how people think about minds generally, including those of other creatures? Research
on nonhuman species has revealed core abilities similar to those in humans, allowing a vari-
ety of creatures to represent objects, quantities, and other animals (Spelke & Kinzler, 2007).
Experiment 3 compared beliefs about human versus animal knowledge.
EXPERIMENT 3
Participants
Two hundred and one adults (M = 33.23 years; 136 female) were recruited through Amazon
Mechanical Turk.
l
D
o
w
n
o
a
d
e
d
f
r
o
m
h
t
t
p
:
/
/
d
i
r
e
c
t
.
m
i
t
.
/
e
d
u
o
p
m
i
/
l
a
r
t
i
c
e
-
p
d
f
/
d
o
i
/
i
/
.
1
0
1
1
6
2
o
p
m
_
a
_
0
0
0
2
8
1
9
2
8
7
8
8
o
p
m
/
/
i
-
0
3
-
8
9
p
d
.
f
b
y
g
u
e
s
t
t
o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3
Figure 2. Mean responses in Experiment 2.
OPEN MIND: Discoveries in Cognitive Science
93
Is Empiricism Innate? Wang, Feigenson
Design, Stimuli, and Coding
We asked participants about core abilities in humans (N = 100) or other animals (N = 101). As
before, we included anchor abilities chosen to elicit agreement that learning was either not re-
quired (seeing in humans/ horses) or was required (washing hands [humans]; using a litter box
[cats]). The critical test items were core perceptual and cognitive abilities demonstrated in both
humans and animals: humans/chickens discriminating faces of conspecifics from nonfaces
(Farroni et al., 2005; Rosa-Salva, Regolin, & Vallortigara, 2010), humans/spiders discriminat-
ing nearby from distant objects (Nagata et al., 2012; Slater et al., 1990), humans/salamanders
discriminating colors (Bornstein et al., 1976; Przyrembel, Keller, & Neumeyer, 1995), humans/
ants discriminating angles of rotation (Landau & Spelke, 1988; Müller & Wehner, 1988),
humans/crows following others’ gaze (Johnson, Slaughter, & Carey, 1998; Schloegl, Kotrschal,
& Bugnyar, 2007), humans/bees discriminating two objects from three (Gross et al., 2009;
Starkey & Cooper, 1980), humans/fish discriminating approximate numerosities (Agrillo, Dadda,
Serena, & Bisazza, 2008; Izard et al., 2009), and humans/chickens thinking a hidden object
still exists (Baillargeon, 1987; Vallortigara, Regolin, Rigoni, & Zanforlin, 1998). Participants
reported whether they believed each ability to be present at birth, and described its origins
using free response. Coding and analysis were as in Experiment 1. Of 2,010 free responses,
18% did not fall into any category and were excluded.
Results
People’s intuitions about the anchor abilities were similar for humans and animals: Adults
reported seeing as present at birth in both species (human M = 98%; horse M = 99%), and
almost never offered learning-based explanations (human M = 1%; horse M = 1%). They
believed that washing hands/using a litter box is not present at birth in either humans or cats
(human M = 0%; cat M = 23%), and offered mostly learning-based explanations (human
M = 100%; cat M = 79%) (binomial exact test ps < .001). Critically, we found that people’s
intuitions diverged for the core ability items: they were significantly less likely to endorse
core abilities as present in newborn humans (M = 37%; 95% CI [32%, 43%]) than newborn
animals (M = 67%; 95% CI [63%, 71%]; t(199) = 8.43, p < .001), and offered more learning-
based explanations for core abilities in humans (M = 62%; 95% CI [56%, 68%]) than animals
(M = 31%; 95% CI[26%, 36%]; t(198) = 7.69, p < .001); this difference was significant for
7/8 core items (χ2s > 4.76, ps < .029) (Figure 3). Whereas people readily explained animals’
abilities by appeal to genes, evolution, and innateness, for the very same abilities in humans
they typically invoked observation and learning.
EXPERIMENT 4
Experiments 1–3 suggest a widespread belief that basic human abilities require learning. That
people did not show this commitment about animals further confirms that this pattern was
not a methodological artifact. But where does this intuitive theory come from? It might form
slowly, perhaps with years of education highlighting learning and instruction. Alternatively, it
might be in place early in life. To find out, in Experiment 4 we tested children.
Participants
Eighty-five children (M = 6.68 years, SD = 1.01; 62 female) were tested at a science museum.
Design, Stimuli, and Coding
Children heard a description of Alex, followed by four examples of Alex’s abilities high-
lighting innateness, maturation, observational learning, and learning through instruction (see
OPEN MIND: Discoveries in Cognitive Science
94
l
D
o
w
n
o
a
d
e
d
f
r
o
m
h
t
t
p
:
/
/
d
i
r
e
c
t
.
m
i
t
.
/
e
d
u
o
p
m
i
/
l
a
r
t
i
c
e
-
p
d
f
/
d
o
i
/
i
/
.
1
0
1
1
6
2
o
p
m
_
a
_
0
0
0
2
8
1
9
2
8
7
8
8
o
p
m
/
/
i
-
0
3
-
8
9
p
d
.
f
b
y
g
u
e
s
t
t
o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3
Is Empiricism Innate? Wang, Feigenson
l
D
o
w
n
o
a
d
e
d
f
r
o
m
h
t
t
p
:
/
/
d
i
r
e
c
t
.
m
i
t
.
/
e
d
u
o
p
m
i
/
l
a
r
t
i
c
e
-
p
d
f
/
d
o
i
/
i
.
/
1
0
1
1
6
2
o
p
m
_
a
_
0
0
0
2
8
1
9
2
8
7
8
8
o
p
m
/
/
i
-
0
3
-
8
9
p
d
.
f
b
y
g
u
e
s
t
t
o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3
Figure 3. Mean responses in Experiment 3.
Experiment 1). Next, children were asked about the abilities from Experiment 1 (two sensory
abilities, seven core knowledge abilities, reading). Items were presented verbally, with props
to illustrate (e.g., cookies with different numbers of chocolate chips to illustrate distinguishing
more from less). Children used the timeline to indicate the age at which they thought abilities
first were present, and were asked “how come” Alex had each ability. If children refused to
answer or gave an uninterpretable response, the experimenter asked a series of follow-up ques-
tions. For example, if children did not answer “How come Alex can see?” the experimenter
asked, “Could Alex see the first time she opened her eyes, or did Alex need to open her eyes
many times to be able to see?” If children answered “many times,” the experimenter asked,
“Did she learn it on her own, or did someone have to teach her?” If children responded that
Alex could do X the first time, this was coded as endorsing innateness. If children said Alex
needed many times before she could X but that no one taught her, this was coded as endorsing
learning through observation. If children said someone taught her, this was coded as endorsing
explicit instruction.
Children’s responses were transcribed and coded. Of 514 responses, 57% did not initially
fall into any defined category, in which case children were asked the follow-up questions.
Results
Children overestimated the onset of core knowledge abilities, citing them as emerging be-
tween 1.27 and 2.01 years (95% CIs [0.99, 1.55] [1.74, 2.29]), more than eight times the
onset suggested by published findings. Children overestimated more for the core abilities than
OPEN MIND: Discoveries in Cognitive Science
95
Is Empiricism Innate? Wang, Feigenson
sensory and reading anchor abilities, t(84) = 10.08, p < .001; t(83) = 5.29, p < .001. Like
adults, children overwhelmingly gave learning-based explanations of the core abilities’ ori-
gins (M = 92%; 95% CI [89%, 95%], t(84) = 25.52, p < .001), rarely invoking innateness
(Figure 4). Indeed, children gave more empiricist explanations of the core abilities than adults
in Experiment 1, t(183) = 5.29, p < .001. This was not due to vocabulary limitations; like
adults, children reported that sensory abilities (seeing and hearing) do not require learning
(M = 24%; 95% CI [17%, 32%]).
EXPERIMENT 5
Experiments 1–4 show that adults and children think of foundational aspects of human knowl-
edge as taking years to emerge and deriving from experience. Is this belief shared by scientists,
including those who study the biological and psychological foundations of the mind? In the
case of the impetus theory, medieval philosopher/scientists held beliefs similar to those of
modern nonspecialist adults. To find out whether contemporary scientists share the com-
mitment that core human abilities are mostly learned, in Experiment 5 we tested professional
academics.
Participants
Participants were 400 faculty, postdocs, and graduate students from 12 geographically diverse
U.S. universities (M = 41.80 years; 272 female): 100 in natural sciences (e.g., chemistry, as-
tronomy), 100 in humanities (e.g., literature, gender studies), and 200 in mind sciences (e.g.,
psychology, neuroscience, linguistics). Thirty-six percent reported having completed a PhD.
l
D
o
w
n
o
a
d
e
d
f
r
o
m
h
t
t
p
:
/
/
d
i
r
e
c
t
.
m
i
t
.
/
e
d
u
o
p
m
i
/
l
a
r
t
i
c
e
-
p
d
f
/
d
o
i
/
i
.
/
1
0
1
1
6
2
o
p
m
_
a
_
0
0
0
2
8
1
9
2
8
7
8
8
o
p
m
/
/
i
-
0
3
-
8
9
p
d
.
f
b
y
g
u
e
s
t
t
o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3
Figure 4. Mean responses in Experiment 4.
OPEN MIND: Discoveries in Cognitive Science
96
Is Empiricism Innate? Wang, Feigenson
Design, Stimuli, and Coding
All aspects were as in Experiment 1. Of 4,000 free responses, 7% did not fall into any defined
category and were excluded.
Results
Academics overestimated the core abilities’ age of onset, citing them as first available between
0.48 and 1.63 years (95% CIs [0.21, 0.74] [1.37, 1.89]), more than four times the age sug-
gested by published findings. Like children and nonspecialist adults, they gave predominantly
learning-based explanations of the abilities’ origins (M = 64%; 95% CI [62%, 67%]) (Figure 5).
However, they offered fewer learning-based explanations than the nonspecialist participants
in Experiment 1, t(497) = 4.76, p < .001, and mind scientists offered fewer learning-based ex-
planations than participants in the natural sciences and humanities, t(399) = 4.32, p < .001.
These results suggest that education may attenuate people’s preference for empiricist explana-
tions. Nonetheless, even mind scientists alone, whose area of expertise included perceptual
l
D
o
w
n
o
a
d
e
d
f
r
o
m
h
t
t
p
:
/
/
d
i
r
e
c
t
.
m
i
t
.
/
e
d
u
o
p
m
i
/
l
a
r
t
i
c
e
-
p
d
f
/
d
o
i
/
i
/
.
1
0
1
1
6
2
o
p
m
_
a
_
0
0
0
2
8
1
9
2
8
7
8
8
o
p
m
/
/
i
-
0
3
-
8
9
p
d
.
f
b
y
g
u
e
s
t
t
o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3
Figure 5. Mean responses in Experiment 5.
OPEN MIND: Discoveries in Cognitive Science
97
Is Empiricism Innate? Wang, Feigenson
and cognitive abilities like those in our survey, preferred to give explanations invoking nurture
over nature (learning-based explanations M = 59%, 95% CI [55%, 63%]).
GENERAL DISCUSSION
For thousands of years humans have pondered the origins of our mental lives, debating the
extent to which aspects of our knowledge rely on nature versus nurture. More recently this
dialogue has moved into the laboratory, with scientists from diverse disciplines seeking to un-
cover what arises in the absence of specific experience versus what must be learned. Our
experiments reveal that naïve adults in the United States and India, children, and highly ed-
ucated adults, including psychologists and neuroscientists, all share the intuition that core
perceptual and cognitive abilities take years to emerge (including abilities for which research
suggests otherwise), and are mostly learned. Whether this widespread belief is right or wrong
is a matter for continued scientific discovery, although in the cases tested here there is com-
pelling evidence that human minds have at least some early structure. Therefore, although a
dichotomous understanding of the nature/nurture debate is surely too simple to be correct (e.g.,
because nature always must anticipate a particular environment in which development will
occur), some of the beliefs of human thinkers appear to run contrary to our current science.
Where does this intuitive theory of human abilities come from? One possibility, to share
a classic quip by psychologist and linguist Lila Gleitman, is that “empiricism is innate.” Al-
though Gleitman may have made the suggestion tongue in cheek, a bias to focus on learning
as the source of knowing could conceivably be a product of evolution, selected for because it
increases pedagogy and encourages information transmission between individuals (Csibra &
Gergely, 2011). Our data are at least consistent with this hypothesis, in that we observed em-
piricist intuitions in the youngest children tested. Alternatively, intuitive empiricism might be
learned—perhaps by noticing the enormous effort and resources humans spend on teaching
(Legare, 2017), by seeing that infants are behaviorally limited, or observing that many human
abilities (like reading) do require experience and practice.1 Such observations could foster the
conclusion that humans lack knowledge until later in life, after experience and instruction have
accrued. If evidence for pedagogy or developmental differences in competence are lacking in
people’s experience of nonhuman animals, this could help explain the divergence in people’s
intuitions about human versus animal abilities. Finally, it is possible that people’s preference
for empiricist explanations is promoted by the feeling that focusing on learning is more opti-
mistic than the alternative. A belief that knowledge is acquired could lead people to conclude
that with relevant experience anything can be learned; this in turn could generate a sense of
equality among individuals. Future work should test these possibilities.
In addition, it will be important for future research to test the scope of intuitive em-
piricism. For aspects of human nature other than the basic perceptual and cognitive abilities
tested here, adults and children often implicate an underlying nature rather than experience—
a phenomenon termed essentialism (Gelman, 2004). For example, people typically believe
that members of categories defined by gender or ethnicity differ in an inherent, unlearned way
1 Noticing that some abilities improve over time might lead participants to report the age at which the ability
is mature, rather than its age of first onset. To steer participants away from responding based the on age of
abilities’ maturity, we always asked when each ability was present “for the first time.” Participants’ systematic
underestimation of the age of reading onset suggests that they were not reporting their beliefs about the age
of asymptotic performance, since reading is an ability that undergoes many years of formal instruction and
improvement.
OPEN MIND: Discoveries in Cognitive Science
98
l
D
o
w
n
o
a
d
e
d
f
r
o
m
h
t
t
p
:
/
/
d
i
r
e
c
t
.
m
i
t
.
/
e
d
u
o
p
m
i
/
l
a
r
t
i
c
e
-
p
d
f
/
d
o
i
/
i
/
.
1
0
1
1
6
2
o
p
m
_
a
_
0
0
0
2
8
1
9
2
8
7
8
8
o
p
m
/
/
i
-
0
3
-
8
9
p
d
.
f
b
y
g
u
e
s
t
t
o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3
Is Empiricism Innate? Wang, Feigenson
(Prentice & Miller, 2007). And when adults are asked about the origins of many human tem-
perament traits (like positive affect), they less often implicate learning and more often appeal
to innate disposition (Haslam, Bastian, & Bissett, 2004). It is against this backdrop that peo-
ple’s commitments about perception and cognition are so remarkable: we seem to believe that
much of what we know derives from experience, but that much of who we are is rooted in
biology.
An inspiration for this research was the recognition that science is held as a paradigm of
inquiry in which evidence—not preference or prejudice—adjudicates between theories. Yet
because science is an endeavor of human minds, and because minds tend to reason about
the world in systematic ways (e.g., Tversky & Kahneman, 1974; Vosniadou & Brewer, 1992),
our intuitive theories unavoidably play a role in forming and evaluating scientific theories
(Bloom & Weisberg, 2007). As such, our findings highlight a key role for psychological research
throughout science: psychology can and should uncover intuitive theories that shape everyday
thought, and that also may affect inquiry across scientific disciplines, from physics to biology
to psychology itself.
ACKNOWLEDGMENTS
We thank the families who participated and the Maryland Science Center. We thank C.
Firestone, S. Gross, and J. Halberda for comments, V. Amandan, K. Carosella, A. McManus, D.
Osaji, M. Shah, and A. Silver for assistance with data collection and coding, and L. Gleitman
for initial inspiration and discussion.
AUTHOR CONTRIBUTIONS
JW: Conceptualization: Equal; Data curation: Lead; Formal analysis: Equal; Methodology:
Equal; Visualization: Equal; Writing - Original Draft: Equal; Writing - Review & Editing: Equal.
LF: Conceptualization: Equal; Formal analysis: Equal; Methodology: Equal; Supervision: Lead;
Visualization: Equal; Writing - Original Draft: Equal; Writing - Review & Editing: Equal.
l
D
o
w
n
o
a
d
e
d
f
r
o
m
h
t
t
p
:
/
/
d
i
r
e
c
t
.
m
i
t
.
/
e
d
u
o
p
m
i
/
l
a
r
t
i
c
e
-
p
d
f
/
d
o
i
/
i
/
.
1
0
1
1
6
2
o
p
m
_
a
_
0
0
0
2
8
1
9
2
8
7
8
8
o
p
m
/
/
i
-
0
3
-
8
9
p
d
.
REFERENCES
Agrillo, C., Dadda, M., Serena, G., & Bisazza, A.
(2008). Do
fish count? Spontaneous discrimination of quantity in female
mosquitofish. Animal Cognition, 11, 495–503. doi:10.1007/
s10071-008-0140-9
Baillargeon, R.
(1987). Object permanence in 3 1/2- and 4 1/2-
month-old infants. Developmental Psychology, 23, 655–664.
Baillargeon, R.
(1995). Physical reasoning in infancy.
In M. S.
Gazzaniga (Ed.), The cognitive neurosciences (pp. 181–204).
Cambridge, MA: MIT Press.
Bloom, P., & Weisberg, D. S. (2007). Childhood origins of adult re-
sistance to science. Science, 316, 996–997.
Bornstein, M. H., Kessen, W., & Weiskopf, S. (1976). Color vision
and hue categorization in young human infants. Journal of Ex-
perimental Psychology: Human Perception and Performance, 2,
115–129. doi:10.1037/0096-1523.2.1.115
Brown, A. M., & Yamamoto, M. (1986). Visual acuity in newborn
and preterm infants measured with grating acuity cards. Amer-
ican Journal of Ophthalmology, 102, 245–253. doi:10.1016/
0002-9394(86)90153-4
f
b
y
g
u
e
s
t
t
o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3
Carey, S.
(2009). The origin of concepts. New York, NY: Oxford
University Press.
Chomsky, N. (1980). Rules and representations. Behavioral and
Brain Sciences, 3, 1–15. doi:10.1017/S0140525X00001515
Cooper, J. M., & Hutchinson, D. S. (1997). Plato: Complete works.
Cambridge, MA: Hackett Publishing.
Csibra, G., & Gergely, G. (2011). Natural pedagogy as evolutionary
adaptation. Philosophical Transactions of the Royal Society B:
Biological Sciences, 366, 1149–1157. doi:10.1098/rstb.2010.0319
Farroni, T., Johnson, M. H., Menon, E., Zulian, L., Faraguna, D.,
& Csibra, G. (2005). Newborns’ preference for face-relevant
stimuli: Effects of contrast polarity. Proceedings of the National
Academy of Sciences, 102, 17245–17250. doi:10.1073/pnas.
0502205102
Gelman, S. A. (2004). Psychological essentialism in children. Trends
in Cognitive Sciences, 8, 404–409. doi:10.1016/j.tics.2004.07.
001
Gross, H. J., Pahl, M., Si, A., Zhu, H., Tautz, J., & Zhang, S.
(2009). Number-based visual generalisation in the honeybee.
PLOS ONE, 4, e4263. doi:10.1371/journal.pone.0004263
OPEN MIND: Discoveries in Cognitive Science
99
Is Empiricism Innate? Wang, Feigenson
Hamlin, J. K., Wynn, K., & Bloom, P. (2007). Social evaluation by
preverbal infants. Nature, 450,557–559. doi:10.1038/nature06288
Hasbrouck, J., & Tindal, G. A. (2011). Oral reading fluency norms:
A valuable assessment tool for reading teachers. The Reading
Teacher, 59, 636–644. doi:10.1598/RT.59.7.3
Haslam, N., Bastian, B., & Bissett, M. (2004). Essentialist beliefs
about personality and their implications. Personality and Social
Psychology Bulletin, 3, 1661–1673.
Izard, V., Sann, C., Spelke, E. S., & Streri, A. (2009). Newborn infants
perceive abstract numbers. Proceedings of the National Academy
of Sciences, 106, 10382–10385. doi:10.1073/pnas.0812142106
Johnson, S., Slaughter, V., & Carey, S. (1998). Whose gaze will in-
fants follow? The elicitation of gaze-following in 12-month-olds.
Developmental Science, 1, 233–238. doi:10.1111/1467-7687.
00036
Landau, B., & Spelke, E. (1988). Geometric complexity and object
search in infancy. Developmental Psychology, 24, 512–521.
Legare, C. H. (2017). Cumulative cultural learning: Development
and diversity. Proceedings of the National Academy of Sciences,
114, 7877–7883. doi:10.1073/pnas.1620743114
Margolis, E., Samuels, R., & Stich, S. P. (2012). The Oxford hand-
book of philosophy of cognitive science. New York, NY: Oxford
University Press.
McCloskey, M. (1983). Intuitive physics. Scientific American, 248,
122–131.
Müller, M., & Wehner, R.
(1988). Path integration in desert ants,
Cataglyphis fortis. Proceedings of the National Academy of Sci-
ences, 85, 5287–5290. doi:10.1073/pnas.85.14.5287
Nagata, T., Koyanagi, M., Tsukamoto, H., Saeki, S.,
Isono, K.,
Shichida, Y., . . . Terakita, A. (2012). Depth perception from im-
age defocus in a jumping spider. Science, 335, 469–471. doi:10.
1126/science.1211667
Northern, J. L., & Downs, M. P. (2002). Hearing in children. New
York, NY: Lippincott Williams & Wilkins.
Prentice, D. A., & Miller, D. T. (2007). Psychological essentialism of
human categories. Current Directions in Psychological Science,
16, 202–206.
Prinz, J. J. (2012). Beyond human nature: How culture and experi-
ence shape our lives. London, UK: Penguin.
Przyrembel, C., Keller, B., & Neumeyer, C. (1995). Trichromatic color
vision in the salamander Salamandra salamandra. Journal of Com-
parative Physiology A, 176, 575–586. doi:10.1007/BF00196422
Rosa-Salva, O., Regolin, L., & Vallortigara, G. (2010). Faces are
special for newly hatched chicks: Evidence for inborn domain-
specific mechanisms underlying spontaneous preferences for
face-like stimuli. Developmental Science, 13, 565–577. doi:10.
1111/j.1467-7687.2009.00914.x
Schloegl, C., Kotrschal, K., & Bugnyar, T. (2007). Gaze following in
common ravens, Corvus corax: Ontogeny and habituation. Animal
Behaviour, 74, 769–778. doi:10.1016/j.anbehav.2006.08.017
Senghas, A., Kita, S., & Özyürek, A. (2004). Children creating core
properties of language: Evidence from an emerging sign language
in Nicaragua. Science, 305, 1779–1782. doi:10.1126/science.
1100199
Skinner, B. F. (2014). Verbal behavior. Cambridge, MA: B. F. Skinner
Foundation.
Slater, A., Mattock, A., & Brown, E. (1990). Size constancy at birth:
Newborn infants’ responses to retinal and real size. Journal of Ex-
perimental Child Psychology, 49, 314–322. doi:10.1016/0022-
0965(90)90061-C
Spelke, E. S., & Kinzler, K. D. (2007). Core knowledge. Developmen-
tal Science, 10, 89–96. doi:10.1111/j.1467-7687.2007.00569.x
Starkey, P., & Cooper, R. G. (1980). Perception of numbers by hu-
man infants. Science, 210, 1033–1035. doi:10.1126/science.
7434014
Tversky, A., & Kahneman, D. (1974). Judgment under uncertainty:
Heuristics and biases. Science, 185, 1124–1131. doi:10.1126/
science.185.4157.1124
Vallortigara, G., Regolin, L., Rigoni, M., & Zanforlin, M. (1998). De-
layed search for a concealed imprinted object in the domestic
chick. Animal Cognition, 1, 17–24. doi:10.1007/s100710050003
Verguts, T., & Fias, W. (2004). Representation of number in animals
and humans: A neural model. Journal of Cognitive Neuroscience,
16, 1493–1504. doi:10.1162/0898929042568497
Vosniadou, S., & Brewer, W. F. (1992). Mental models of the earth: A
study of conceptual change in childhood. Cognitive Psychology,
24, 535–585. doi:10.1016/0010-0285(92)90018-W
Wang, J., & Feigenson, L. (2019). Supplemental material for “Is
empiricism innate? Preference for nurture over nature in
people’s beliefs about the origins of human knowledge.” Open
Mind: Discoveries in Cognitive Science, 3. doi:10.1162/opmi_a_
00028
OPEN MIND: Discoveries in Cognitive Science
100
l
D
o
w
n
o
a
d
e
d
f
r
o
m
h
t
t
p
:
/
/
d
i
r
e
c
t
.
m
i
t
.
/
e
d
u
o
p
m
i
/
l
a
r
t
i
c
e
-
p
d
f
/
d
o
i
/
i
.
/
1
0
1
1
6
2
o
p
m
_
a
_
0
0
0
2
8
1
9
2
8
7
8
8
o
p
m
/
/
i
-
0
3
-
8
9
p
d
.
f
b
y
g
u
e
s
t
t
o
n
0
7
S
e
p
e
m
b
e
r
2
0
2
3
