Psychological Science
http://pss.sagepub.com/
Heralding the Authoritarian? Orientation Toward Authority in Early Childhood
Michal Reifen Tagar, Christopher M. Federico, Kristen E. Lyons, Steven Ludeke and Melissa A. Koenig
Psychological Science 2014 25: 883 originally published online 6 February 2014
DOI: 10.1177/0956797613516470
The online version of this article can be found at:
http://pss.sagepub.com/content/25/4/883
Published by:
http://www.sagepublications.com
On behalf of:
Association for Psychological Science
Additional services and information for Psychological Science can be found at:
Email Alerts: http://pss.sagepub.com/cgi/alerts
Subscriptions: http://pss.sagepub.com/subscriptions
Reprints: http://www.sagepub.com/journalsReprints.nav
Permissions: http://www.sagepub.com/journalsPermissions.nav
>> Version of Record - Apr 9, 2014
OnlineFirst Version of Record - Feb 6, 2014
What is This?
Downloaded from pss.sagepub.com at COLGATE UNIV LIBRARY on June 29, 2014
516470
research-article2014
PSSXXX10.1177/0956797613516470Reifen Tagar et al.Heralding the Authoritarian
Research Article
Heralding the Authoritarian? Orientation
Toward Authority in Early Childhood
Psychological Science
2014, Vol. 25(4) 883­–892
© The Author(s) 2014
Reprints and permissions:
sagepub.com/journalsPermissions.nav
DOI: 10.1177/0956797613516470
pss.sagepub.com
Michal Reifen Tagar1, Christopher M. Federico1,2, Kristen E. Lyons3,
Steven Ludeke4, and Melissa A. Koenig5
1
Department of Psychology, University of Minnesota; 2Department of Political Science, University of
Minnesota; 3Department of Psychology, Metropolitan State University of Denver; 4Department of
Psychology, Colgate University; and 5Institute of Child Development, University of Minnesota
Abstract
In the research reported here, we examined whether individual differences in authoritarianism have expressions in
early childhood. We expected that young children would be more responsive to cues of deviance and status to the
extent that their parents endorsed authoritarian values. Using a sample of 43 preschoolers and their parents, we found
support for both expectations. Children of parents high in authoritarianism trusted adults who adhered to convention
(vs. adults who did not) more than did children of parents low in authoritarianism. Furthermore, compared with
children of parents low in authoritarianism, children of parents high in authoritarianism gave greater weight to a statusbased “adult = reliable” heuristic in trusting an ambiguously conventional adult. Findings were consistent using two
different measures of parents’ authoritarian values. These findings demonstrate that children’s trust-related behaviors
vary reliably with their parents’ orientations toward authority and convention, and suggest that individual differences
in authoritarianism express themselves well before early adulthood.
Keywords
authoritarianism, selective trust, epistemic trust, parent-child correspondence, early childhood, individual differences,
values, personal values, heuristics, learning, childhood development
Received 5/15/13; Revision accepted 11/14/13
Do individual differences associated with authoritarianism express themselves in early childhood? Traditional
accounts suggest that they do not; rather, authoritarian
values are assumed to crystallize in early adulthood along
with other sociopolitical orientations (e.g., Altemeyer,
1998). Departing from this perspective, we argue that
individual differences reflecting authoritarian tendencies
exist well before early adulthood and that, among young
children, these individual differences are expressed as a
greater responsiveness to cues of status and of deviance
when determining whom to learn from. Specifically, we
predicted that children of parents high (vs. low) in
authoritarian values should (a) be more discriminating in
trusting (i.e., choosing to learn from) adults who had
previously demonstrated conventional word labeling versus adults who had used blatantly nonconventional word
labels and (b) give greater weight to a status-based
“adults are to be trusted” heuristic in trusting an ambiguously conventional adult. To examine these predictions,
we looked at authoritarianism-consistent patterns of
deference to convention and authority in early childhood
in a sample of 3- to 4-year-olds and their parents. In
the following sections, we outline our theory and
predictions.
The Origins of Authoritarianism
A long line of research shows that adults who are high
in authoritarianism—a general tendency to submit
to established authority and social convention—show
greater ethnocentrism, xenophobia, political intolerance,
and prejudice (Adorno, Frenkel-Brunswik, Levinson, &
Sanford, 1950; Altemeyer, 1996, 1998; Duckitt, 2001; Jost,
Corresponding Author:
Michal Reifen Tagar, Department of Psychology, University of
Minnesota, 75 East River Rd., Minneapolis, MN 55455
E-mail: [email protected]
Downloaded from pss.sagepub.com at COLGATE UNIV LIBRARY on June 29, 2014
Reifen Tagar et al.
884
Glaser, Kruglanski, & Sulloway, 2003; Stenner, 2005). It is
not surprising, then, that researchers concerned with
sociopolitical orientations have long been interested in
the origins of authoritarianism. The pioneering work of
Adorno et al. (1950) proposed a psychodynamic explanation based on children’s repressed and projected hostility
toward their strict and punitive parents. In light of weak
empirical support for this account and a general loss of
interest in psychodynamic approaches (for a review, see
Brown, 1965), later researchers—most notably, Altemeyer
(1988, 1996)—have argued for a social-learning explanation. This perspective points to a consistently strong correlation between young adults’ and their parents’ levels
of authoritarianism (e.g., Altemeyer, 1996, 1998; see also
Duriez & Soenens, 2009; Peterson, Smirles, & Wentworth,
1997). From a very different perspective, research in
behavior genetics has also highlighted within-family consistency in levels of authoritarianism among adults, with
twin studies suggesting a considerable level of heritability in authoritarianism (Ludeke & Krueger, 2013; McCourt,
Bouchard, Lykken, Tellegen, & Keyes, 1999).
Despite this consistent interest in the genesis of authoritarian tendencies and parent-child similarities in authoritarianism, research on the topic has rarely examined or
even anticipated early-childhood manifestations of authoritarianism. In part, this limited interest in childhood expressions of authoritarianism is due to early failures to find
support for Adorno et al.’s (1950) psychodynamic hypotheses about the childhood origins of authoritarianism
(Brown, 1965). In addition, it can be traced to a more
recent tendency to conceptualize authoritarianism as a
learned sociopolitical orientation that—like other political
predispositions (Niemi & Hepburn, 1995)—does not crystallize prior to early adulthood (Altemeyer, 1981, 1998; see
also Sears & Levy, 2003).
However, the existence of early-childhood individual
differences in the manifestations of authoritarianism may
have been unduly dismissed. As noted previously, research
indicates that authoritarianism may have a substantial
genetic component, which raises the possibility that
the authoritarian “phenotype” may have early-childhood
expressions that cannot be captured by measures intended
for adults. Moreover, researchers have begun to reconceptualize authoritarianism as a prepolitical psychological tendency to defer to authority and social convention,
independent of its sociopolitical expressions and consequences and motivated by values emphasizing social conformity (Federico, Fisher, & Deason, 2011; Feldman, 2003;
Stenner, 2005). Both of these developments leave room to
wonder if individual differences are in fact apparent before
early adulthood. Our goal in the present study was
to explore this question by examining whether behavior
indicative of a psychological predisposition toward
authoritarianism consistent with parental values can be
seen as early as the preschool years.
Authoritarianism and Selective Trust
But how might we detect early manifestations of authoritarianism in children that parallel their parents’ values?
We believe that recent developmental research on
“selective trust” in children provides a promising avenue. According to this literature, although children rely
on others for much of their early learning (Csibra &
Gergely, 2011), they are not indiscriminately credulous
and are motivated to gain reliable, culturally appropriate knowledge (Koenig & Sabbagh, 2013). Accordingly,
early on, children develop strategies for discriminating
between different sources of information based on cues
of subjective source relevance (Harris & Corriveau,
2011). Two strategies in particular stand out. First, children are more likely to trust new information provided
by sources who reliably apply conventional, socially
accepted labels to objects (Koenig, Clement, & Harris,
2004; Koenig & Harris, 2005). Second, children are more
likely to trust new information from sources seen as
having status, particularly adults ( Jaswal, 2004; Taylor,
Cartwright, & Bowden, 1991). These status cues are
overridden when sources fail to reliably apply conventional labels ( Jaswal & Neely, 2006). Indeed, theoretical
models suggest that these cues of reliability and status
may have deep roots in evolved social-learning mechanisms common to most human societies (Heinrich &
McElreath, 2003).
Thus far, attentiveness to these two cues has been
understood as a normative developmental process, without regard to individual differences. However, in adults,
deference to convention and to those considered high in
status are consistently stronger among those high in
authoritarianism (Adorno et al., 1950; Altemeyer, 1996).
Thus, to the extent that the authoritarian phenotype manifests itself in childhood in ways that parallel parental
orientations, we argue that young children of authoritarian parents should rely more strongly on the aforementioned cues of reliability and status. Specifically, we
hypothesized (a) that children of parents who are high
(vs. low) in authoritarianism and related social-conformity values should be more sensitive to cues of conventionality—that is, they should trust adults who reliably
apply conventional labels more than adults who do not
to a greater extent than do children of parents low
in authoritarianism—and (b) that children of parents high
(vs. low) in authoritarianism and related social-conformity
values should be more sensitive to adult status—that is,
they should be more likely to trust an adult speaker in
the absence of clear cues of conventionality. We tested
Downloaded from pss.sagepub.com at COLGATE UNIV LIBRARY on June 29, 2014
Heralding the Authoritarian885
these hypotheses in a sample of preschoolers and their
parents.
After parents gave informed consent, children joined the
experimenter in a child-friendly space to complete a
selective-trust game and a verbal-intelligence test while
parents completed a survey in the back of the room.
After the parents and children had completed these measures, they were thanked for their participation, and children received a small toy as a gift.
watched four short video clips of an unfamiliar adult
labeling common objects (e.g., Fig. 1a).
In the second, “test” phase, children watched four
short video clips of the same speaker introducing novel
objects (e.g., unusual kitchen gadgets) and labeling them
(e.g., “modi,” as in Fig. 1b). After each new label was
given, children were presented with an image of the
novel object (in the absence of the speaker) and were
asked whether the label provided by the speaker was the
correct label for the object: “She says that’s a(n) X—what
do you think, is that a(n) X?” Three iterations of this game
were played. In each, children viewed a different speaker
in the familiarization phase: (a) one who labeled each of
four familiar objects using conventional terms (e.g., labeling a shoe a “shoe”; conventional speaker); (b) one who
labeled two of the four familiar objects conventionally
and the other two unconventionally (e.g., calling a shoe
a “ball”; ambiguous speaker); and (c) one who labeled all
four objects unconventionally (unconventional speaker).1
Each of the speakers labeled different common and novel
objects. In describing this game, it is important to emphasize that the adult speakers were strangers to the children
and that they never interacted with them in person; they
were seen only on video. Moreover, when the children
gave their responses, neither the speaker nor an image of
the speaker was present. This minimized the possibility
that the children were merely demonstrating external
compliance with authority as opposed to epistemic trust.2
Selective-trust game. The selective-trust game had
two parts. In the first, “familiarization” phase, children
Measures. Intercorrelations between key variables are
shown in Table 1.
Method
Participants
Participants were 3- and 4-year-old children (21 girls, 22
boys; mean age = 3.99 years, SD = 5.47 months; age
range = 37–59 months) and one parent of each. Three
participants were dropped: One was not a native English
speaker, 1 did not want to play, and 1 was a twin of an
earlier participant. Approximately 90% of the sample was
Caucasian and 10% was of mixed ethnic backgrounds. All
participants were of upper-middle-class socioeconomic
status. Families were recruited from a database maintained by the University of Minnesota’s Institute of Child
Development.
Procedure, materials, and measures
Fig. 1. Examples of (a) familiarization-phase and (b) test-phase stimuli.
Downloaded from pss.sagepub.com at COLGATE UNIV LIBRARY on June 29, 2014
Reifen Tagar et al.
886
Table 1. Intercorrelations Between Key Study Variables
Variable
1.
2.
3.
4.
5.
6.
7.
Verbal IQ
Gender
Parental authoritarianism
Parental social-conformity values
Number of trusting responses to unconventional speaker
Number of trusting responses to conventional speaker
Number of trusting responses to ambiguous speaker
1
2
—
–.11
–.23
–.34*
–.05
–.22
–.27†
—
–.38*
–.15
.16
–.10
.08
3
—
.60***
–.18
.26
.27†
4
—
–.11
.28†
.32*
5
6
—
.27†
.22
—
.54***
†
p < .10. *p < .05. ***p < .001.
Children’s verbal intelligence. To control for variation in children’s verbal intelligence, we administered
the verbal subtest of the Stanford-Binet Intelligence Scales
for Early Childhood (Roid, 2003), which asks children
to label objects or actions, describe images of situations, and define words. Scores reflect the number of
points earned on this task based on a standardized rubric
(M = 16.41, SD = 4.61).
Children’s gender. Child gender was dummy-coded
(0 = male, 1 = female).
Parents’ authoritarian predisposition. Parents were
asked to indicate in four forced-choice items which of
two child-rearing values (authoritarian and nonauthoritarian) they found more important (Feldman & Stenner,
1997; Stenner, 2005). The value pairs were “independence” versus “respect for elders,” “obedience” versus
“self-reliance,” “curiosity” versus “good manners,” and
“being considerate” versus “[being] well-behaved.” After
scoring (authoritarian response = 1, nonauthoritarian
response = 0), responses were summed and then recoded
to run from 0 to 1, with higher scores indicating a stronger authoritarian predisposition (M = .39, SD = .28).
Parents’ social-conformity values. Recent work has
suggested that authoritarianism may be conceptualized
as a generalized motive for social conformity (Feldman,
2003; Stenner, 2005). Therefore, we also administered a
measure of social-conformity values to provide convergent evidence for any authoritarianism-related findings.
Parents were asked to indicate in 17 forced-choice items
which of two statements—one prioritizing social conformity (scored 1) and one prioritizing personal autonomy
(scored 0)—they agreed with more (example pair: “Society should aim to protect citizens’ right to live any way
they choose” vs. “It is important to enforce the community’s standards of right and wrong”). Responses were
summed and then recoded to run from 0 to 1, with higher
scores indicating greater conformity (M = .36, SD = .17).
Results
Parental orientations, speaker
conventionality, and children’s
trusting responses
Our first hypothesis is that the tendency for children
to selectively trust conventional speakers more than
unconventional ones will be stronger among children
whose parents are higher in authoritarianism and socialconformity values. To examine this prediction, we used a
multilevel modeling approach. We treated the counts of
trusting responses for the conventional speaker and for
the unconventional speaker as separate observations,
which gave us two observations per participant. Within
this structure, speaker conventionality was represented
using a dummy variable (0 = unconventional, 1 = conventional). Because the dependent measure was the
number of trusting responses across four trials for each of
the two speakers, mixed-model binomial logistic regression was used to analyze the data.3 To reflect the nonindependence of observations within each participant,
the intercept was allowed to randomly vary across participants. Because our hypothesis assumes that the impact
of speaker conventionality should vary across participants as a function of parental orientations, we also
allowed the slope for speaker conventionality and the
covariance between this slope and the intercept to vary
randomly.
We first looked at trust as a function of speaker conventionality and parental authoritarianism. Two models
were estimated. In the first, trust was regressed on the
speaker-conventionality dummy and parental authoritarianism, along with two controls (verbal IQ and gender).
In the second model, the product term for the interaction
between conventionality and authoritarianism was added.
Both authoritarianism and verbal IQ were mean-centered.
The results are summarized in Table 2.
As expected, the variance component for the intercept
was significantly different from zero in both models,
Downloaded from pss.sagepub.com at COLGATE UNIV LIBRARY on June 29, 2014
Heralding the Authoritarian887
Table 2. Results From Mixed-Model Binomial Logistic Regressions Predicting
the Number of Trusting Responses From Parental Authoritarianism and Speaker
Conventionality
Predictor
Model 1
Fixed effects
Verbal IQ
Gender
Speaker conventionality
Parental authoritarianism
Authoritarianism × Speaker Conventionality
Intercept
Random effects
σ2 (intercept)
σ2 (speaker conventionality)
σ (intercept, speaker conventionality)
–2 log-likelihood
–0.09 (0.13)
0.21 (1.19)
4.48*** (1.30)
–1.27 (2.53)
—
–1.37 (0.85)
8.62 (4.49)
18.63 (10.17)
–1.77 (5.76)
206.13
Model 2
–0.11 (0.13)
0.21 (1.18)
4.21*** (1.09)
–2.66 (2.40)
7.16* (3.45)
–1.39† (0.84)
8.99 (4.44)
15.23 (7.51)
–2.91 (4.51)
201.40
Note: Values shown are unstandardized coefficients; the dependent variable is the number
of trusting responses across four trials per within-subjects speaker condition. Standard errors
are shown in parentheses. There were two responses (one per speaker) for each of 40
participants, for a total of 80 Level 1 units and 40 Level 2 units. Gender was coded as 0 for
male and 1 for female; speaker conventionality was coded as 0 for unconventional and 1 for
conventional.
†
p < .10. *p < .05. ***p < .001.
authoritarianism, the effect of speaker conventionality
was much stronger (b = 6.22, p < .001), amounting to an
increase of .87 in the probability of a trusting response.
Thus, although children of parents both low and high in
Unconventional Speaker
Conventional Speaker
1.0
Probability of Trusting Response
according to likelihood-ratio tests—Model 1: χ2(1) =
77.60; Model 2: χ2(1) = 76.73; both ps < .001. Further
likelihood-ratio tests indicated that the addition of random effects for the speaker-conventionality slope and the
covariance between this slope and the intercept also
improved model fit over and above an intercept-only
random-effect specification—Model 1: χ2(2) = 38.16;
Model 2: χ2(2) = 31.79; both ps < .001.
Looking at the fixed effects in Model 1, we found only
a main effect for speaker conventionality (b = 4.48, p <
.001), such that children gave more trusting responses in
reaction to the conventional speaker than the unconventional speaker (e.g., Koenig et al., 2004). More importantly, in Model 2, we found the predicted interaction
between parental authoritarianism and speaker conventionality (b = 7.16, p < .05). To break this down, we
looked at the simple slopes for speaker conventionality 1
standard deviation below the mean and 1 standard deviation above the mean of parental authoritarianism (Aiken
& West, 1991). To make these estimates more comprehensible, we also computed predicted probabilities of
a trusting response for each combination of the two
parental-authoritarianism levels and the two speakers
(see Fig. 2).
Among children of parents low in authoritarianism,
trust was marginally higher for the conventional speaker
than the unconventional speaker (b = 2.21, p < .10),
amounting to an increase of .49 in the probability of a
trusting response. Among children of parents high in
.8
.6
.4
.2
.0
Low Parental
Authoritarianism
High Parental
Authoritarianism
Fig. 2. Predicted probability of a child’s trusting response as a function
of parental authoritarianism and speaker conventionality. Predictions
are based on estimates shown in Table 2.
Downloaded from pss.sagepub.com at COLGATE UNIV LIBRARY on June 29, 2014
Reifen Tagar et al.
888
authoritarianism trusted a conventional speaker more
than an unconventional speaker, the effect was particularly pronounced for children of parents high in authoritarianism, as expected.
To replicate this finding, we repeated our analysis using
a second authoritarianism-related measure: Feldman’s
(2003) measure of social-conformity values. Model specification and estimation was identical to that reported above,
except that conformity was substituted for authoritarianism. The results are summarized in Table 3.
Again, likelihood-ratio tests on the variance component for the intercept were significantly different from
zero in both models—Model 1: χ2(1) = 77.31; Model 2:
χ2(1) = 77.93; both ps < .001. Moreover, the addition of
random effects for the speaker-conventionality slope and
the covariance between this slope and the intercept
improved model fit over and above an intercept-only
random-effect specification—Model 1: χ2(2) = 38.10;
Model 2: χ2(2) = 32.52; both ps < .001.
Turning to the fixed effects, we found only a main
effect of speaker conventionality in Model 1 (b = 4.50,
p < .001); as before, children were more trusting in reaction to the conventional speaker. In Model 2, we found
the predicted interaction between speaker conventionality and parental social-conformity values (b = 11.15, p <
.05). To unpack this, we looked at the simple slopes for
speaker conventionality 1 standard deviation below the
mean (low conformity) and 1 standard deviation above
the mean (high conformity) of parental social-conformity
values. We also computed predicted probabilities of a trusting response for each combination of the two parentalconformity levels and the two speakers (see Fig. 3).
Among children of parents low in social-conformity
values, trust was higher in response to the conventional
speaker than the unconventional speaker (b = 2.62, p <
.05), for an increase of .56 in the probability of a trusting
response. Among children of parents high in social-conformity values, the impact of speaker conventionality was
even stronger (b = 6.37, p < .001), for an increase of .8 in
the probability of a trusting response. Thus, although
children of parents both low and high in social-conformity values responded to the conventional speaker with
greater trust, the effect of speaker conventionality was
larger for children of parents high in social-conformity
values, as expected.
Trust in ambiguously conventional
speakers as a function of parental
orientations
Our second hypothesis suggests that children of highly
authoritarian or conformity-valuing parents will place
more trust in an adult speaker whose level of conventionality is ambiguous than will children of parents who are
low in authoritarianism or social-conformity values. To
examine this, we looked at the relationship between
parental authoritarianism and social-conformity values and
trusting responses to the third speaker, who modeled conventional responses prior to half of the trust trials and
unconventional responses prior to the other half. This trust
count was regressed on each parental variable in a separate model, with verbal IQ and gender as controls. Binomial
logistic regression was used to estimate the models. Results
for authoritarianism are shown in the top half of Table 4.
Table 3. Results From Mixed-Model Binomial Logistic Regressions Predicting the
Number of Trusting Responses From Parental Social-Conformity Values and Speaker
Conventionality
Predictor
Model 1
Fixed effects
Verbal IQ
Gender
Speaker conventionality
Parental social-conformity values
Social Conformity × Speaker Conventionality
Intercept
Random effects
σ2 (intercept)
σ2 (speaker conventionality)
σ (intercept, speaker conventionality)
–2 log-likelihood
–0.10 (0.13)
0.36 (1.12)
4.50*** (1.32)
–1.91 (4.09)
—
–1.49† (0.80)
8.70 (4.50)
18.57 (10.20)
–1.74 (5.96)
206.16
Model 2
–0.12 (0.13)
0.30 (1.14)
4.49*** (1.18)
–4.14 (3.84)
11.15* (5.39)
–1.52† (0.82)
9.20 (4.52)
15.51 (7.84)
–2.48 (4.55)
201.38
Note: Values shown are unstandardized coefficients; the dependent variable is the number of
trusting responses across four trials per within-subjects speaker condition. Standard errors are
shown in parentheses. There were two responses (one per speaker) for each of 40 participants,
for a total of 80 Level 1 units and 40 Level 2 units. Gender was coded as 0 for male and 1 for
female; speaker conventionality was coded as 0 for unconventional and 1 for conventional.
†
p < .10. *p < .05. ***p < .001.
Downloaded from pss.sagepub.com at COLGATE UNIV LIBRARY on June 29, 2014
Heralding the Authoritarian889
value of authoritarianism (Long & Freese, 2005). This
estimate indicated that such a move was associated with
an increase of .43 in the probability of a trusting response
(see Fig. 4a).
Results for parental social-conformity values are shown
in the bottom half of Table 4. Like authoritarianism,
parental social conformity was associated with greater
trust in response to the ambiguously conventional
speaker (b = 2.97, p < .01), after the effects of the control
variables were taken into account. For illustration, we
also computed a first difference for parental conformity.
This estimate indicated that moving from the lowest to
the highest level of parental-conformity values was associated with an increase of .51 in the probability of a trusting response (see Fig. 4b). Thus, we found support for
our second hypothesis: The more authoritarian a child’s
parent is, the greater weight the child will attribute to a
speaker’s adult status.
Unconventional Speaker
Conventional Speaker
Probability of Trusting Response
1.0
.8
.6
.4
.2
.0
Low Parental
Conformity
High Parental
Conformity
Fig. 3. Predicted probability of a child’s trusting response as a function of parental social-conformity values and speaker conventionality.
Predictions are based on estimates shown in Table 3.
Greater parental authoritarianism was associated with
a higher number of trusting responses to the ambiguously conventional adult speaker (b = 1.85, p < .01), after
the effects of the control variables were taken into
account. For clarity, we also computed a first difference
for parental authoritarianism, which indicated the simulated change in the probability of a trusting response
associated with moving from the lowest to the highest
Table 4. Results of Binomial Logistic Regressions Predicting
Trusting Responses to an Ambiguously Conventional Speaker
From Parental Authoritarianism and Social-Conformity Values
b
Predictor
Authoritarianism model
Verbal IQ
Gender
Parental authoritarianism
Intercept
–2 log-likelihood
–0.09† (0.04)
0.60 (0.37)
1.84** (0.69)
–0.41 (0.26)
157.21
Social-conformity model
Verbal IQ
Gender
Parental social-conformity values
Intercept
–2 log-likelihood
–0.06† (0.04)
0.37 (0.34)
2.96** (1.10)
–0.27 (0.25)
156.97
Note: The dependent variable is the number of trusting responses
across four trials (N = 40). Standard errors are shown in parentheses.
Gender was coded as 0 = male, 1 = female.
†
p < .10. **p < .01.
Discussion
We examined whether individual differences in behavior
congruent with authoritarianism—and with parental values—are expressed in the preschool years. In a sample of
3- and 4-year-olds, we found that children of parents high
(vs. low) in authoritarianism and social-conformity values
were (a) more likely to trust conventional (vs. unconventional) adult speakers and (b) more likely to exhibit statusbased trust in adults in the absence of clear cues of
conventionality. That is, children of authoritarian parents
gave greater weight to cues of both conventionality and
(adult) status when determining the trustworthiness of a
speaker. Given the established association between parents’ and their adult children’s levels of authoritarianism,
we suggest that these results reveal emerging manifestations of authoritarianism at an early age.4
In addition to being the earliest demonstration of such
individual differences we are aware of, this work also
contributes to the existing literature in several other ways.
First, in contrast to prior work suggesting that the development of selective trust is a normative, age-dependent
phenomenon that occurs in all children, our results indicate that there are systematic individual differences in
selective trust—and that these differences can be reliably
predicted from the psychological characteristics of children’s parents.
Second, it suggests that parents’ authoritarianism predicts not merely their children’s attitudes but also more
subtle differences in children’s trust behaviors in the context of learning—and at a far younger age than classical
work suggests that authoritarianism should manifest
itself. Specifically, children of authoritarian parents were
more likely to demonstrate an “authoritarian” pattern of
rejecting information from nonconventional sources and
Downloaded from pss.sagepub.com at COLGATE UNIV LIBRARY on June 29, 2014
Reifen Tagar et al.
890
b
1.0
Probability of Trusting Response
Probability of Trusting Response
a
.8
.6
.4
.2
.0
–.4 –.3 –.2 –.1
.0
.1
.2
.3
.4
.5
Parental Authoritarianism (mean-centered)
.6
1.0
.8
.6
.4
.2
.0
–.3
–.2
–.1
.0
.1
.2
.3
.4
Parental Conformity (mean-centered)
Fig. 4. Predicted probability of a child’s trusting response to an ambiguously conventional speaker as a function of (a) parental authoritarianism
and (b) parental social-conformity values. Predictions are based on estimates shown in Table 4; the shaded bands represent 95% confidence
intervals.
accepting information from conventional and highstatus (i.e., adult) sources. This seems to foreshadow
both the lack of openness and the greater reliance on
heuristic processing shown by authoritarians in adulthood (Kemmelmeier, 2009). This is all the more impressive in light of the fact that the sources were strangers
who had no relationship with the children, never interacted with the children, and were not present even in the
form of an image when children gave their responses.
Finally, our research can also be distinguished from
work on individual differences in children’s willingness
to internalize their caregivers’ moral agendas (e.g.,
Kochanska, Koenig, Barry, Kim, & Yoon, 2010). Rather
than focusing on the internalization of moral rules, we
examined the relationship between parental values and
children’s willingness to learn from other adults in an
epistemic sense. Further work should examine how parents express these values and how such expressions
might shape children’s learning in both the epistemic and
the moral realms.
More generally, the source of differences in selective
trust and their association with variables such as parental
authoritarianism merits further study. On the one hand,
these differences might stem from genetic commonality
between parents and children, as implied by work on the
heritability of authoritarianism (Ludeke & Krueger, 2013;
McCourt et al., 1999). On the other hand, other work has
suggested that genetically inherited tendencies in the
related domain of political conservatism do not emerge
until after children leave home, with environmental
forces being the main ideological influence prior to this
(Hatemi et al., 2009). This, in contrast, suggests that
our findings might reflect socialization. Of course, our
findings might reflect a combination of both forces.
Finally, greater discrimination between conventional and
unconventional sources and greater deference to adults in
the absence of clear conventionality cues could also reflect
mechanisms that contribute to the long-term development
and expression of authoritarianism in adults. That is, early
differences in learning style—whether mediated by genes
or socialization—may predispose some children more
than others to accept “truths” from those who accurately
adhere to convention or possess status, producing a cascade of experiences that encourage the emergence of the
typical adult authoritarian phenotype. This suggests that
authoritarianism may result not merely from what children
learn throughout their childhood but also whom they
agree to learn from.
Author Contributions
M. Reifen Tagar developed the study concept and led the project.
All authors contributed to the study design. Testing and data
collection were performed by K. E. Lyons and M. A. Koenig.
M. Reifen Tagar performed all initial analyses, and C. M. Federico
performed all advanced analyses. M. Reifen Tagar, C. M.
Federico, and M. A. Koenig drafted the manuscript, and S.
Ludeke and K. E. Lyons provided critical revisions. All authors
approved the final version of the manuscript for submission.
Acknowledgments
We thank Dr. Sanford Weisberg for his advice on statistical
analyses.
Declaration of Conflicting Interests
The authors declared that they had no conflicts of interest with
respect to their authorship or the publication of this article.
Downloaded from pss.sagepub.com at COLGATE UNIV LIBRARY on June 29, 2014
Heralding the Authoritarian891
Notes
1. Note that inaccurate labelers simultaneously deviate from the
conventions of a given language community and are factually
wrong in their statements. However, statements in language
are true only because words within the statements are used
conventionally: Word meanings do not have truth status independent of their conventional status (Lewis, 1969). Therefore,
accuracy violations are necessarily violations of convention.
Thus, our labeling of the responses as conventional or unconventional represents the notion that such responses are correct
or incorrect vis-à-vis linguistic convention.
2. Two versions of the videos were created, with different adults
serving as the different speakers (i.e., in Version A, the speaker
wearing a green shirt was the conventional speaker, whereas
in Version B, she was the unconventional speaker); video version was counterbalanced between subjects. The speakers were
also presented in one of two orders: conventional, ambiguous,
unconventional or unconventional, ambiguous, conventional.
Inclusion of a dummy variable corresponding to the order
manipulation in the models in Tables 2 and 3 did not change
the key findings with respect to the interactions between parental orientations and speaker accuracy, so we omit the results of
this model here for the sake of parsimony. These results are
available upon request from the authors.
3. Binomial logistic regression is a variant of logistic regression
used when the outcome is not “success” or “failure” on a single
dichotomous binomial trial but, rather, a count or proportion
of the number of successes across N binomial trials (Cook &
Weisberg, 1999). In the present context, each decision by a
child to trust or not trust an adult speaker in the test phase constituted a binomial trial, with four total trials in each condition;
each trust response is counted as a “success.”
4. As numerous studies have demonstrated, authoritarianism is
correlated with political conservatism (e.g., Altemeyer, 1996).
To be sure that authoritarianism was not confounded with conservatism, we administered a 7-point scale of parental conservatism (from 1, very liberal, to 7, very conservative). We reran
the models shown in Tables 2 and 3 with conservatism and
a Conservatism × Speaker interaction included. These models
revealed neither a significant main effect nor an interaction
involving conservatism, and the key interactions between the
authoritarianism variables and speaker condition remained significant. Similarly, we reran the models shown in Table 4 with
the main effect of conservatism added. The coefficients for conservatism in these models were nonsignificant, and the coefficients for the authoritarianism variables remained significant.
Thus, our key effects are robust.
References
Adorno, T. W., Frenkel-Brunswik, E., Levinson, D., & Sanford,
R. N. (1950). The authoritarian personality. New York, NY:
Harper.
Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing
and interpreting interactions. Newbury Park, CA: Sage.
Altemeyer, B. (1981). Right-wing authoritarianism. Winnipeg,
Canada: The University of Manitoba Press.
Altemeyer, B. (1988). Enemies of freedom: Understanding rightwing authoritarianism. San Francisco, CA: Jossey-Bass.
Altemeyer, B. (1996). The authoritarian specter. Cambridge,
MA: Harvard University Press.
Altemeyer, B. (1998). The other “authoritarian personality.” In
M. P. Zanna (Ed.), Advances in experimental social psychology (Vol. 30, pp. 47–92). San Diego, CA: Academic Press.
Brown, R. (1965). Social psychology. London, England: Collier
Macmillan.
Cook, R. D., & Weisberg, S. (1999). Applied regression including computing and graphics. New York, NY: John Wiley
& Sons.
Csibra, C., & Gergely, G. (2011). Natural pedagogy as evolutionary adaptation. Philosophical Transactions of the Royal
Society B: Biological Sciences, 366, 1149–1157.
Duckitt, J. (2001). A dual-process cognitive-motivational theory
of ideology and prejudice. In M. P. Zanna (Ed.), Advances
in experimental social psychology (Vol. 33, pp. 41–113). San
Diego, CA: Academic Press.
Duriez, B., & Soenens, B. (2009). The intergenerational transmission of racism: The role of right-wing authoritarianism
and social dominance orientation. Journal of Research in
Personality, 43, 906–909.
Federico, C. M., Fisher, E. L., & Deason, G. (2011). Political expertise and the link between the authoritarian predisposition
and conservatism. Public Opinion Quarterly, 75, 686–708.
Feldman, S. (2003). Enforcing social conformity: A theory of
authoritarianism. Political Psychology, 24, 41–74.
Feldman, S., & Stenner, K. (1997). Perceived threat and authoritarianism. Political Psychology, 4, 741–770.
Harris, P. L., & Corriveau, K. H. (2011). Young children’s selective trust in informants. Philosophical Transactions of the
Royal Society B: Biological Sciences, 366, 1179–1187.
Hatemi, P. K., Funk, C., Medland, S. E., Maes, H., Silberg, J.,
Martin, N. G., & Eaves, L. J. (2009). Genetic and environmental transmission of political attitudes over a life time.
The Journal of Politics, 71, 1141–1156.
Heinrich, J., & McElreath, R. (2003). The evolution of cultural
evolution. Evolutionary Anthropology, 12, 123–135.
Jaswal, V. K. (2004). Don’t believe everything you hear:
Preschoolers’ sensitivity to speaker intent in category
induction. Child Development, 75, 1871–1885.
Jaswal, V. K., & Neely, L. A. (2006). Adults don’t always know
best: Preschoolers use past reliability over age when learning new words. Psychological Science, 17, 757–758.
Jost, J. T., Glaser, J., Kruglanski, A. W., & Sulloway, E. (2003).
Political conservatism as motivated social cognition.
Psychological Bulletin, 129, 339–375.
Kemmelmeier, M. (2009). Authoritarianism and its relationship
with intuitive-experiential cognitive style and heuristic processing. Personality and Individual Differences, 48, 44–48.
Kochanska, G., Koenig, J. L., Barry, R. A., Kim, S., & Yoon, J. E.
(2010). Children’s conscience during toddler and preschool
years, moral self, and a competent, adaptive developmental
trajectory. Developmental Psychology, 46, 1320–1332.
Koenig, M. A., Clement, F., & Harris, P. L. (2004). Trust in
testimony: Children’s use of true and false statements.
Psychological Science, 15, 694–698.
Koenig, M. A., & Harris, P. L. (2005). Preschoolers mistrust
ignorant and inaccurate speakers. Child Development, 76,
1261–1277.
Downloaded from pss.sagepub.com at COLGATE UNIV LIBRARY on June 29, 2014
Reifen Tagar et al.
892
Koenig, M. A., & Sabbagh, M. A. (2013). Selective social learning: New perspectives on learning from others [Introduction
to Special Issue: Selective Social Learning]. Developmental
Psychology, 49, 399–403.
Lewis, D. (1969). Convention: A philosophical study. Cambridge,
MA: Harvard University Press.
Long, J. S., & Freese, J. (2005). Regression models for categorical
outcomes using Stata (2nd ed.). College Station, TX: Stata
Press.
Ludeke, S. G., & Krueger, R. F. (2013). Authoritarianism as a
personality trait: Evidence from a longitudinal behavior
genetic study. Personality and Individual Differences, 55,
480–484.
McCourt, K., Bouchard, T. J., Jr., Lykken, D. T., Tellegen, A., &
Keyes, M. (1999). Authoritarianism revisited: Genetic and
environmental influences examined in twins reared apart
and together. Personality and Individual Differences, 27,
985–1014.
Niemi, R. G., & Hepburn, M. A. (1995). The rebirth of political
socialization. Perspectives on Political Science, 24, 7–10.
Peterson, B. E., Smirles, K. A., & Wentworth, P. A. (1997).
Generativity and authoritarianism: Implications for personality, political involvement, and parenting. Journal of
Personality and Social Psychology, 72, 1202–1216.
Roid, G. H. (2003). Stanford-Binet Intelligence Scales, Fifth
Edition. Itasca, IL: Riverside.
Sears, D. O., & Levy, S. (2003). Childhood and adult political
development. In D. O. Sears, L. Huddy, & R. Jervis (Eds.),
Oxford handbook of political psychology (pp. 60–109).
Oxford, England: Oxford University Press.
Stenner, K. (2005). The authoritarian dynamic. Cambridge,
England: Cambridge University Press.
Taylor, M., Cartwright, B. S., & Bowden, T. (1991). Perspective
taking and theory of mind: Do children predict interpretive
diversity as a function of differences in observers’ knowledge? Child Development, 62, 1334–1351.
Downloaded from pss.sagepub.com at COLGATE UNIV LIBRARY on June 29, 2014