You are on page 1of 7

Learning and Individual Differences 40 (2015) 156162

Contents lists available at ScienceDirect

Learning and Individual Differences

journal homepage:

Able, ready, and willing: Examining the additive and interactive effects of
intelligence, conscientiousness, and autonomous motivation on
undergraduate academic performance
Stefano I. Di Domenico , Marc A. Fournier
University of Toronto Scarborough, Canada

a r t i c l e i n f o a b s t r a c t

Article history: Intelligence, conscientiousness, and autonomous motivation are well-established predictors of academic perfor-
Received 10 January 2014 mance. However, research has yet to examine how these variables combine and interact in the prediction of
Received in revised form 22 January 2015 academic success. We therefore examined intelligence, conscientiousness, and autonomous motivation in the
Accepted 29 March 2015
concurrent prediction of students' grade point average (GPA) among university undergraduates. Conscientious-
ness was a stronger predictor of GPA at higher levels of intelligence, suggesting that an industrious disposition
serves a catalytic function among those students who are the most intellectually able. Conscientiousness was a
Conscientiousness stronger predictor of GPA at lower levels of autonomous motivation, suggesting that an industrious disposition
Autonomous motivation also serves a compensatory function among those students who are the least intrinsically interested. These
Self-determination theory ndings call for further research on Intelligence Conscientiousness and Conscientiousness Autonomous
Academic performance Motivation interactions in the prediction of academic performance.
2015 Elsevier Inc. All rights reserved.

1. Introduction = .23 for conscientiousness, and = .16 for autonomous motivation

(corrected for measurement error; Richardson, Abraham, & Bond,
Academic success can play a dramatic role in shaping the life course 2012). However, as these individual difference characteristics have
of university students. Apart from being an important admission re- been studied in relative isolation, very little is still known about how
quirement for graduate and professional schools, excellent performance these variables combine and interact in the prediction of academic
in undergraduate courses also promises better job opportunities after performance in university settings. The purpose of the present re-
university (Plant, Ericsson, Hill, & Asberg, 2005; Strenze, 2007). A natu- search was thus to examine the additive and potentially interactive
ral turn of interest for education researchers is thus to identify what (i.e., synergistic or compensatory) effects of intelligence, conscientious-
individual difference characteristics, either alone or in combination, ness, and autonomous motivation on grade point average (GPA) among
are most predictive of academic success. a sample of university undergraduates.
Findings from three separate and largely independent lines of
research have identied three sources of individual differences 1.1. A closer look at intelligence: the ability to do academic work
that contribute to the prediction of academic performance. Under-
graduate students differ in terms of (a) their ability to do academic Intelligence refers to the general mental capability that subsumes
work (i.e., intelligence), (b) their readiness to do academic work a broad range of more specic cognitive abilities, including abstract
(i.e., conscientiousness), and (c) their willingness to do academic reasoning, planning, problem-solving, and learning from experience
work (i.e., autonomous motivation). Meta-analytic effect sizes for (Gottfredson, 1997). Spearman (1904, 1927) was the rst to posit that
these predictors have been estimated to be = .21 for intelligence, a single common factor could be responsible for producing the positive
manifold of correlations among tests of specic mental abilities, a posi-
tion that has garnered continued support in subsequent investigations
Author Note: As both authors contributed equally to the preparation of this article,
names are listed in alphabetical order. Portions of these ndings were presented at the (Carroll, 1993; Jensen, 1998). The heritability of general intelligence
2013 biennial meeting of the Association for Research in Personality (ARP) in Charlotte, increases with age, from about 30% in childhood to about 80% in adult-
North Carolina. We express our appreciation to Zhouran (Crystal) Li, Aubrey Gibson, and hood (Deary, Penke, & Johnson, 2010). Individual differences in general
Mesaaba Correia for their contributions to the collection of these data. intelligence are very stable over time; for example, Deary, Whalley,
Corresponding author at: Department of Psychology, University of Toronto
Scarborough, 1265 Military Trail, Toronto, Ontario M1C 1A4, Canada.
Lemmon, Crawford, and Starr (2000) tested participants at age 11 and
E-mail addresses: (S.I. Di Domenico), then again at age 79 and found that general intelligence had a rank- (M.A. Fournier). order stability coefcient of .63.
1041-6080/ 2015 Elsevier Inc. All rights reserved.
S.I. Di Domenico, M.A. Fournier / Learning and Individual Differences 40 (2015) 156162 157

Although intelligence is thought to reect a capability for learning abiding sense of self or, stated in attributional terms, as having an inter-
and comprehension that is broader and deeper than mere book- nal perceived locus of causality (Ryan & Connell, 1989). SDT differenti-
learning or test-taking abilities (Gottfredson, 1997), academic perfor- ates behavioral regulations along a continuum of relative autonomy.
mance has traditionally been used an important criterion to validate The most basic differentiation in this regard concerns the difference be-
psychometric tests of intelligence. Indeed, Simon and Binet developed tween intrinsic motivation and extrinsic motivation. Intrinsic motivation
the rst intelligence tests over 100 years ago specically to help identify refers to the impetus for behavior performed for the inherent satisfac-
children that would have difculties learning in a regular classroom tion associated with its enactment (i.e., for interest's sake). In contrast,
environment. The distinct cognitive abilities subsumed under the gen- extrinsic motivation refers to the impetus for behavior aimed at the at-
eral intelligence factor may each account for additional variance on tainment of instrumentally separable outcomes (i.e., for the attainment
tests restricted to specic academic subjects (e.g., processing speed of rewards or the avoidance of punishments). SDT further distinguishes
and spatial ability in the domain of mathematics; Lubinski, Webb, between three empirically distinct forms of extrinsic motivation.
Morelock, & Benbow, 2001; Rohde & Thompson, 2007). However, Identied regulation is a relatively autonomous type of extrinsic motiva-
researchers have mostly focused on the role of general intelligence in tion that is evidenced when one performs an activity because one
the prediction of academic performance as broadly dened by course recognizes or accepts the activity's importance or underlying value.
grades or GPA, a convention followed in the present investigation. Introjected regulation is a less autonomous type of extrinsic motivation
that is evidenced when one performs an activity to avoid feelings
1.2 . A closer look at conscientiousness: the readiness to do academic work of shame and guilt or to defensively maintain feelings of self-worth.
External regulation is the least autonomous type of extrinsic motivation
Conscientiousness is an individual difference construct in the ve- that is evidenced when one performs an activity to obtain rewards or
factor model of personality, a taxonomy of personality traits that also avoid punishments.
includes neuroticism, extraversion, openness to experience, and agree- Unlike intelligence and conscientiousness, which are highly de-
ableness (Goldberg, 1993; John, Naumann, & Soto, 2008; McCrae & contextualized and heritable attributes, autonomous motivation is a
Costa, 2008). Roberts, Jackson, Fayard, Edmonds, and Meints (2009, context-specic characteristic that is believed to have its primary basis
p. 369) described conscientiousness as the propensity to follow socially in those environments in which the relevant goal-directed behaviors
prescribed norms for impulse control, to be goal directed, to plan, and to are socialized and enacted (Deci & Ryan, 2000, 2012; Ryan & Deci,
be able to delay gratication and follow norms and rules. Conscien- 2008). In the language of the ve-factor theory of personality (McCrae
tiousness was discovered in factor-analytic studies of personality- & Costa, 2008), autonomous motivation is considered a characteristic
descriptive terms in natural languages. Like the other ve-factor traits, adaptation, an individual difference characteristic that is jointly deter-
conscientiousness has a strong genetic basis; a number of studies esti- mined by one's foundational personality traits and by one's environ-
mate its heritability to be about 50% in adulthood (e.g., Jang, Livesley, mental context, although the latter set of inuences is believed to
& Vemon, 1996; Jang, McCrae, Angleitner, Riemann, & Livesley, 1998; predominate in the case of autonomous motivation. This makes auton-
Loehlin, McCrae, Costa, & John, 1998). Although people's levels of omous motivation a malleable characteristic. Indeed, a very large body
conscientiousness tend to increase throughout the lifespan (Roberts, of applied work in SDT shows that socializing agents (e.g., parents,
Walton, & Viechtbauer, 2006), individual differences in conscientious- educators, workplace supervisors, etc.) play a critical role in fostering
ness are stable during the typical university years; for example, Roberts, the development of autonomous motivation by encouraging people's
Caspi, and Moft (2001) tested participants at age 18 and then again initiation, by providing them with meaningful choices, by offering
at age 26 and found that conscientiousness had a rank-order stability them structured and task-relevant feedback, and by making them feel
coefcient of .67. valued. The benets of such autonomy supportive practices have been
Much like intelligence, conscientiousness represents a broad con- documented across a wide variety of life domains, including educational
struct that accounts for the variance shared by more narrow or specic settings (Deci & Ryan, 2000, 2012; Ryan & Deci, 2008).
personality facets. Recently, researchers have identied ve principal The earliest applications of SDT in the educational domain focused
facets to the trait domain of conscientiousness (Roberts, Bogg, Walton, on primary and secondary school settings. These studies found that
Chernyshenko, & Stark, 2004; Roberts, Chernyshenko, Stark, & the more autonomously motivated students tended to exhibit a variety
Goldberg, 2005): industriousness (achievement vs. laziness), orderliness of positive outcomes, including better maintenance and transference of
(organization vs. sloppiness), impulse control (cautiousness vs. care- academic activities, greater conceptual understanding, less procrastina-
lessness), reliability (dependability vs. unreliability), and formality tion, lower drop-out rates, and higher levels of achievement (see Deci &
(traditionalism vs. nonconformity). Although researchers have mostly Ryan, 2000). Although fewer studies have applied SDT to university
focused on the broad trait domain dened by conscientiousness, emerg- settings, researchers have found that more autonomous forms of moti-
ing studies suggest that the facets of conscientiousness are differentially vation are similarly benecial at higher levels of education (e.g., Black &
associated with academic performance, with the achievement-oriented Deci, 2000; Komarraju, Karau, & Schmeck, 2009; Kusurkar, Ten Cate,
facets (i.e., industriousness and reliability) being the strongest predictors Vos, Westers, & Croiset, 2012; Miquelon, Vallerand, Grouzet, &
in university settings (e.g., Noftle & Robins, 2007; Paunonen & Ashton, Cardinal, 2005; Ning & Downing, 2012; Vansteenkiste, Sierens,
2013). In the present study, we examined both conscientiousness and Soenens, Luyckx, & Lens, 2009). However, researchers have yet to exam-
its constituent facets alongside general intelligence and autonomous ine whether autonomous motivation has meaningful incremental utility
motivation in the prediction of academic performance. over intelligence and conscientiousness in the prediction of undergrad-
uate academic performance.
1.3. A closer look at autonomous motivation: the willingness to do academic
work 1.4 . Overview of the present research

Autonomous motivation is a key construct in a macro-theoretical In the present research, we examined the additive and potentially
framework for the study of motivation and personality development interactive effects of intelligence, conscientiousness, and autonomous
called self-determination theory (SDT; Deci & Ryan, 2000, 2012; Ryan motivation to the prediction of undergraduate academic performance.
& Deci, 2008). Autonomous motivation refers to the extent to which Given that these constructs each represent a qualitatively distinct
people experience their goal-relevant behaviors as being choicefully ini- class of individual differences, we expected the correlations among
tiated, volitionally enacted, and personally endorsed. Autonomously their measures to range from minimal (intelligence and conscientious-
motivated behaviors are thus experienced as emanating from one's ness) to moderate (conscientiousness and autonomous motivation).
158 S.I. Di Domenico, M.A. Fournier / Learning and Individual Differences 40 (2015) 156162

We accordingly hypothesized that each of these variables would con- to predict academic performance across a variety of educational con-
tribute incrementally to the prediction of undergraduate academic texts (Wonderlic, Inc., 2002). Respondents are presented 50 test items
performance. of increasing difculty and are instructed to answer as many of these
We also considered the possibility that intelligence, conscientious- items as possible within a 12-minute period with each correct answer
ness, and autonomous motivation could combine non-additively in counting for one point. Scores on the WPTTM converge with other
the prediction of academic performance. On the one hand, we imagined well-established measures of general intelligence such as full-scale IQ
that these variables could combine in a synergistic (i.e., more-than- as assessed by the Wechsler Adult Intelligence Scale, Third Edition
additive) manner, wherein each variable would have its greatest effect (i.e., .75 b r b .96; Wonderlic, Inc., 2002, p. 39). The descriptive statistics
on academic performance at higher levels of the other variables. Follow- for the WPTTM are presented in Table 1.
ing Cohen, Cohen, West, and Aiken (2003), we call this a synergistic
interaction. On the other hand, we imagined that these variables could
2.2.2. Trait conscientiousness
also combine in a compensatory (i.e., less-than-additive) manner,
We administered 48 items from the Behavioral Indicators of
wherein each variable would have its greatest effect on academic
Conscientiousness (BIC; Jackson et al., 2010) that pertained to the ve
performance at lower levels of the other variables. Following Cohen
principal facets of conscientiousness (Roberts et al., 2005,). We thus
et al. (2003), we call this a compensatory interaction.
administered a 9-item subscale for industriousness, an 18-item subscale
Hypotheses concerning synergistic interactions between intelli-
for orderliness, a 9-item subscale for impulse control, a 6-item subscale
gence and constructs related to the personality trait domain of consci-
for reliability, and a 6-item subscale for formality (see the Appendix of
entiousness were popular for some time in the eld of industrial/
Jackson et al., 2010). Participants were asked to indicate how often
organizational psychology (Maier, 1955). Although research has
they perform specic conscientious behaviors on a scale ranging from
historically supported the more straightforward additive model
1 (Never) to 5 (Quite often). Sample items include Finish a set amount
(Sackett, Gruys, & Ellingson, 1998), recent studies have found syner-
of work before relaxing (industriousness), Set a timeline for getting
gistic interaction effects between intelligence and the achievement
a project done (orderliness), Cancel or switch plans at the last minute
(i.e., industriousness) facet of conscientiousness in the prediction of
(impulse control; reverse-scored), Oversleep for class or work
job performance (Perry, Hunter, Witt, & Harris, 2010). Perry et al.
(reliability; reverse-scored), and Call someone Ms., Mr., Mrs., Sir, etc.
(2010) suggested that individuals who value achievement may be
(formality). A conscientiousness total score on the BIC was computed
more motivated to deploy their cognitive abilities in the service of
by standardizing and summing participants' facet-level scores. The
job performance. Although intelligence and conscientiousness have
descriptive statistics for the BIC facet scales are presented in Table 1.
typically been considered independent predictors in the domain of aca-
Given the heterogeneity of item content across the facet measures, the
demic performance (e.g., DeYoung, 2011), we know of no investigation
internal consistency of this Conscientiousness composite was relatively
that has explicitly tested their interaction. Accordingly, we followed
lower than that of its constituent facet scales, = .60.
Perry et al. (2010) and hypothesized that intelligence and conscien-
tiousness would combine synergistically in the prediction of academic
performance. We further hypothesized that intelligence and the indus- 2.2.3. Autonomous motivation
triousness facet of conscientiousness would combine synergistically The Academic Self-Regulation Scale (ASRS; Vansteenkiste et al.,
in the prediction of academic performance. Given that we had neither 2009) presents students with a 16-item list of reasons for studying
theoretical nor empirical grounds on which to formally hypothesize university course materials. Four items each correspond to intrinsic
whether autonomous motivation would interact with either intelli- motivation, identied regulation, introjected regulation, and external
gence or conscientiousness, these two-way interactions were tested regulation. Sample items include I study because it's fun (intrinsic
on an exploratory basis. motivation), I study because I want to learn new things (identied
regulation), I study because I want others to think I'm a good student
2. Method (introjected regulation), and I study because that's what others
(e.g., parents, friends) expect me to do (external regulation). Partici-
2.1. Participants pants were instructed to rate the personal importance of each reason
on a scale ranging from 1 (Completely not important) to 7 (Very important).
A total of 271 undergraduate students (58% female) from the The descriptive statistics for the ASRS scales are presented in Table 1.
University of Toronto Scarborough (UTSC) participated in a 1-hour lab-
oratory session, either for course credit in their introductory psychology Table 1
course or for monetary compensation of $15 CAN. Participants con- Descriptive statistics of measures.
vened for the current study either by themselves or in unacquainted
M SD Number of scale items
two-, three-, or four-person groups. The measures that were utilized
in the current study were administered in the same sequence that General intelligence
WPTTM 23.94 5.81
they are listed below. Participants ranged in age from 19 to 29 (M =
Facets of conscientiousness
20.83, SD = 1.67), although ve students did not report their age. The BIC-Industriousness 3.59 .69 .82 9
sample was academically diverse, including students in their rst BIC-Orderliness 3.14 .81 .90 18
(11%), second (39%), third (21%), fourth (21%), and fth or above (6%) BIC-Impulse control 3.04 .65 .76 9
year of studies. The vast majority of students (89%) had decided upon BIC-Reliability 4.20 .64 .77 6
BIC-Formality 3.24 .70 .68 6
a program of study. These students varied greatly in terms of their
Motivation to study
degree programs and their academic histories were mostly character- ASRS-Intrinsic motivation 4.45 1.47 .89 4
ized by an interdisciplinary prole of course combinations across the ASRS-Identied regulation 5.80 1.08 .77 4
sciences, social sciences, and humanities. ASRS-Introjected regulation 4.34 1.51 .78 4
ASRS-External regulation 4.13 1.56 .83 4
Academic performance
2.2. Measures Cumulative GPA 2.61 .82

Note: N = 271; aN = 266; WPTTM = Wonderlic Personnel TestTM; BIC = Behavioral

2.2.1. General intelligence Indicators of Conscientiousness; ASRS = Academic Self-Regulation Scale; GPA = grade
The Wonderlic Personnel TestTM (WPTTM) is a brief, reliable, and point average; Item responses on the BIC and ASRS were obtained on 5- and 7-point
valid measure of general intelligence that has been extensively utilized Likert-type scales, respectively. All statistics are rounded to the nearest hundredth.
S.I. Di Domenico, M.A. Fournier / Learning and Individual Differences 40 (2015) 156162 159

Table 2 with cumulative GPA (although the correlation with autonomous

Correlations among the four subscales of the ASRS (below the diagonal) and 95% motivation was only marginally statistically signicant). Students' year
condence intervals (above the diagonal).
of study was not associated with intelligence, conscientiousness, or au-
1 2 3 4 tonomous motivation. However, students' year of study was positively
1. Intrinsic motivation [.52, .67] [.11, .13] [.30, .07] associated with cumulative GPA, suggesting the presence of grade ina-
2. Identied regulation .60 [.09, .32] [.27, .04] tion (Johnson, 2003) in higher years of undergraduate studies. We
3. Introjected regulation .01 .21 [.40, .58] accordingly statistically controlled for the effect of students' academic
4. External regulation .19 .16 .50
year of study in all subsequent analyses.
Note: N = 271; p b .001, p b .01, p b .05; ASRS = Academic Self-Regulation Scale;
the simplex-like pattern of correlations along the main diagonal is bolded. All statistics 3.2. Regression analyses
are rounded to the nearest hundredth.

We utilized a step-up regression procedure to examine the additive

Table 2 shows that the different types of motivation evidenced a and interactive effects of intelligence, conscientiousness, and autono-
simplex-like pattern of associations, such that the largest correla- mous motivation on academic performance. Step 1 featured the main
tions appeared along the main diagonal of the matrix (Ryan & Connell, effects for intelligence, conscientiousness, and academic year of study.
1989). Following previous studies (e.g., Black & Deci, 2000; Weinstein Autonomous motivation was entered at Step 2 in order to test its incre-
& Ryan, 2010), we accordingly operationalized autonomous motivation mental predictive validity over intelligence and conscientiousness. Step
by scores on a relative autonomy index (RAI), which were computed 3 featured the two-way interaction effects between (a) intelligence and
by the following formula: RAI = intrinsic motivation + identied conscientiousness, (b) intelligence and autonomous motivation, and
regulation introjected regulation external regulation. (c) conscientiousness and autonomous motivation. We standardized
each of our predictors but left GPA scores unstandardized so that the
2.2.4. Academic performance lower-order terms in our models could more straightforwardly be
At the very end of the study, participants downloaded and printed interpreted as GPA points per standard deviation unit.
an unofcial copy of their transcript from the undergraduate Student As expected, the results at Step 1 of our hierarchical procedure were
Web Services at the University of Toronto ( All consistent with the idea that intelligence and conscientiousness operate
personally identifying information was removed prior to analyses. We as independent predictors of academic performance. Autonomous
utilized students' cumulative GPA as our criterion variable. Of the 271 motivation, entered at Step 2, did not evince incremental predictive util-
students who participated in the study, ve were unable to provide ity over and above intelligence and conscientiousness. A partial F-test
their transcripts because of difculties in accessing their online student comparing the models at Steps 1 and 2 formally supported this conclu-
account. sion, R2 = .01, F(1, 261) = 1.58, and p = .210. The nal model at Step 3
indicated a signicant Intelligence Conscientiousness interaction and
2.3. Data analytic approach a signicant Conscientiousness Autonomous Motivation interaction.
The Intelligence Autonomous Motivation interaction was not signi-
We conducted two types of statistical analyses. We rst conducted cant (Table 4). A partial F-test found that the model at Step 3 explained
correlational analyses to examine if the well-established relationships signicantly more variance than the model at Step 2, R2 = .03,
that intelligence, conscientiousness, and autonomous motivation have F(3, 258) = 3.37, and p = .019.
with academic performance would replicate in the current sample. We formally examined the signicant Intelligence Conscientiousness
We then conducted a hierarchical or step-up multiple regression interaction by plotting and testing the simple slopes of intelligence
procedure (Aiken & West, 1991) to examine the additive and interactive at high (+ 1 SD) and low ( 1 SD) levels of conscientiousness
effects of intelligence, conscientiousness, and autonomous motivation (Aiken & West, 1991). This interaction is illustrated in Fig. 1. The simple
on GPA. Partial F-tests were utilized to examine whether progressively effect of intelligence was greater at high levels of conscientiousness,
complex models explained signicantly greater amounts of variance b = .40, SE = .06, t = 6.16, and p b .001, than at low levels of conscien-
in GPA. We conducted an analogous set of analyses to examine the tiousness, b = .17, SE = .07, t = 2.63, and p = .009. To further examine
role of the conscientiousness facets in the prediction of academic perfor- the signicant Intelligence Conscientiousness interaction, we also
mance. The results and discussion of these analyses are detailed in the plotted and tested the simple slopes of conscientiousness at high
Supplemental materials. (+ 1 SD) and low ( 1 SD) levels of intelligence. The simple effect of
conscientiousness was greater at high levels of intelligence, b = .20,
3. Results SE = .07, t = 2.80, and p = .006, than at low levels of intelligence, for
which the simple effect was not signicantly different than zero,
3.1. Correlational analyses b = .03, SE = .06, t = .49, and p = .623. The patterning of this
interaction is thus consistent with the synergistic model.
Table 3 displays the correlations between all of the primary vari- We similarly examined the signicant Conscientiousness
ables. In keeping with previous research, intelligence, conscientious- Autonomous Motivation interaction by plotting and testing the simple
ness, and autonomous motivation were each positively associated slopes of autonomous motivation at low ( 1 SD) and high (+ 1 SD)

Table 3
Correlations among the primary variables (below the diagonal) and 95% condence intervals (above the diagonal).

1 2 3 4 5 6 7

1. Agea [.06, .30] [.63, .75] [.18, .06] [.18, .06] [.00, .24] [.15, .10]b
2. Sex .18 [.11, .13] [.04, .20] [.22, .01] [.10, .14] [.18, .06]
3. Year of study .70 .01 [.03, .20] [.12, .11] [.07, .17] [.02, .26]
4. Intelligence .06 .08 .09 [.05, .19] [.08, .16] [.28, .49]
5. Conscientiousness .06 .11 .01 .07 [.18, .40] [.02, .26]
6. Autonomous motivation .12 .02 .05 .04 .30 [.00, .23]
7. Cumulative GPAa .02b .06 .15 .39 .14 .12

Note: N = 271; aN = 266; bN = 261; p b .01, p b .05, p b .10, Sex: Female = 1, Male = 1; GPA = grade point average; all the listed correlations are rounded to the nearest hundredth.
160 S.I. Di Domenico, M.A. Fournier / Learning and Individual Differences 40 (2015) 156162

Table 4
Regression analyses for the prediction of cumulative GPA (N = 266).

Step Predictors b(SE) 95% CI p-value Model summary

Step 1 Year of study .09 (.05) [.00, .19] .042 R2 = .18, F(3, 262) = 18.84,
p b .001
Intelligence .30 (.05) [.21, .40] b .001
Conscientiousness .09 (.05) [.00, .18] .048
Step 2 Autonomous Motivation .06 (.05) [.03, .15] .210 R2 = .18, F(4, 261) = 14.55,
p b .001
Step 3 Intelligence Conscientiousness .11 (.05) [.02, .21] .020 R2 = .21, F(7, 258) = 9.99,
p b .001
Intelligence Autonomous Motivation .00 (.05) [.10, .10] .986
Conscientiousness Autonomous Motivation .09 (.04) [.17, .01] .037

Note: The effect size b(SE) is expressed in units of cumulative GPA per standard deviation along the predictor. The p-values are rounded to the nearest thousandth. All other statistics are
rounded to the nearest hundredth.
95% CI = 95% condence interval.

levels of conscientiousness (Aiken & West, 1991). This interaction is 4.1. General Intelligence and Trait Conscientiousness
illustrated in Fig. 2. The simple effect of autonomous motivation was
greater at low levels of conscientiousness, b = .17, SE = .06, t = 2.65, The results of the present study indicated that intelligence and
and p = .008, than at high levels of conscientiousness, for which conscientiousness are not independent predictors of academic perfor-
the simple effect was not signicantly different than zero, b = .00, mance. Rather, these variables proved to synergistically operate with
SE = .06, t = .07, and p = .950. To further examine the signicant each other in the prediction of academic performance. Specically, we
Conscientiousness Autonomous Motivation interaction, we also plot- found that the contributions of intelligence to academic performance
ted and tested the simple slopes of conscientiousness at low (1 SD) were amplied by conscientiousness. It thus appears that conscien-
and high (+1 SD) levels of autonomous motivation. The simple effect tiousness facilitates or catalyzes some or all of the processes through
of conscientiousness was greater at low levels of autonomous motiva- which intelligence inuences academic performance. One possibility is
tion, b = .17, SE = .06, t = 2.79, and p = .006, than at high levels of that brighter students are able to better understand their learning mate-
autonomous motivation, for which the simple effect was not signicant- rial, and brighter students who bring their intellectual talent to bear on
ly different than zero, b = .00, SE = .07, t = .07, and p = .950. The pat- their studies through hard work (i.e., those higher in conscientiousness)
terning of this interaction is thus consistent with the compensatory stand to obtain the highest levels of academic performance.
model. Another possibility concerns the underlying neurophysiological sys-
tems, particularly those of the prefrontal cortex, through which intelli-
gence enhances people's capacity for self-regulation (e.g., Shamosh
4. Discussion et al., 2008) and the role of conscientiousness in promoting the deploy-
ment of those systems (e.g., Pailing & Segalowitz, 2004). In a recent
The present research examined the additive and interactive effects study (Di Domenico, Rodrigo, Ayaz, Fournier, & Ruocco, in press), we
of intelligence, conscientiousness, and autonomous motivation in the found that brighter individuals evidenced neural efciency (i.e., lower
prediction of undergraduate academic performance. Data from an aca- levels of neural activation and faster reaction times) in a region of the
demically diverse sample of university students revealed a synergistic prefrontal cortex called the right inferior frontal gyrus (rIFG) during a
interaction between intelligence and conscientiousness and a compen- decision-making task that had previously implicated activity in the an-
satory interaction between conscientiousness and autonomous motiva- terior cingulate cortex (ACC) as reecting the degree of choice difculty
tion. That is, conscientiousness performed a catalytic function among (Nakao et al., 2010). While the ACC is characterized as an evaluative
students with higher levels of intelligence and a compensatory function system, signaling the need for greater attentional resources (Yeung,
among students with lower levels of autonomous motivation. We Botvinick, & Cohen, 2004), the rIFG is characterized as playing an exec-
discuss each of these ndings in greater detail below. utive role, overseeing the wilful selection and control of behavior

Fig. 1. Predicted cumulative GPA across levels of intelligence and conscientiousness. Error Fig. 2. Predicted cumulative GPA across levels of conscientiousness and autonomous
bars represent 90% condence intervals. motivation. Error bars represent 90% condence intervals.
S.I. Di Domenico, M.A. Fournier / Learning and Individual Differences 40 (2015) 156162 161

through the inhibition of competing behavioral outputs (Aron, Robbins, 4.3. Limitations
& Poldrack, 2004). Indeed, activity in the rIFG observed during con-
trolled laboratory studies is a predictor of successful self-control in The current study is not without its limitations. The rst limitation
ecologically valid settings (Berkman, Falk, & Lieberman, 2011). Notably, concerns the relatively exploratory nature of the present study. Intelli-
more conscientious individuals exhibit a consistently pronounced ACC gence, conscientiousness, and autonomous motivation are well-
response (e.g., Pailing & Segalowitz, 2004), suggesting that they more established predictors of academic performance in university but
reliably engage the cognitive-control mechanisms of the rIFG when previous studies have not examined these constructs simultaneously
the necessity strikes. Thus, one possibility is that the synergistic interac- and we accordingly considered different possibilities for how these
tion between intelligence and conscientiousness in the prediction of variables could combine and interact in the prediction of academic suc-
academic performance has less to do with learning per se and more cess. Although the presently documented interactions suggest interest-
to do with the reliable deployment and the efcient functioning of ing avenues for future studies examining these constructs, a crucial step
people's self-regulatory systems. Research should continue to test for for future research will be to replicate the present ndings using larger
synergistic interactions between intelligence and conscientiousness in samples. The second limitation concerns the delimited population from
the prediction of academic and non-academic performance outcomes. which the present sample was drawn. Although the present sample was
academically diverse and involved students of various ages, all partici-
pants came from the same institution. Generalizations based on the cur-
4.2. Autonomous motivation and trait conscientiousness rent nding should therefore be made with caution. The third limitation
concerns our global assessment of autonomous motivation. It is possible
Because the interaction between conscientiousness and autono- that university students are more or less autonomously motivated for
mous motivation was examined on an exploratory basis, the results of each of the different courses in which they are enrolled. With our global
the current study only offer preliminary support for a compensatory assessment of autonomous motivation, it is unclear how this possibility
interaction between these two constructs. Nevertheless, the current may have played itself out in the current investigation. Future research
results do suggest potentially fruitful avenues for future studies. Specif- could address this issue by adopting a more idiographic approach in
ically, the compensatory interaction may signify that the processes predicting academic performance, for example, through the use of mul-
through which autonomous motivation inuences academic perfor- tilevel models (Snijders & Bosker, 2012) specied to examine possible
mance overlap with those of conscientiousness and that these processes cross-level interactions between person-level trait conscientiousness
can operate to enhance performance outcomes only to a limited extent. and course-level autonomous motivation. The fourth limitation con-
A natural step for future research is to therefore examine the basis of cerns the cross-sectional design of our study. Although the models
this compensatory interaction by testing alternative models of mediat- examined in the present investigation were based upon past research
ed moderation with different candidate mediator variables. Indeed, a ndings that have separately established intelligence, conscientious-
variety of constructs have been associated with conscientiousness, ness, and autonomous motivation to each be important predictors of
autonomous motivation, and academic performance. One such con- academic performance, the performance criterion that was utilized in
struct is academic effort, which has been previously established as the present study was assessed concurrently with the predictors. Future
both a mediator of the relationship between conscientiousness and research should therefore attempt to recover the interaction effects
academic performance (e.g., Trautwein, Ldtke, Roberts, Schnyder, & observed in the present study by utilizing longitudinal designs for the
Niggli, 2009) and as a typical outcome of autonomous motivation prospective examination of academic performance.
(e.g., Deci & Ryan, 2000). Academic effort is a good candidate mediator
because there is likely to be a limit on the extent to which incremental 4.4. Conclusion
increases of effort will yield incremental increases of academic perfor-
mance and on the extent to which people can exert effort in the rst Three separate and largely independent lines of research suggest
place. that students must be able (intelligent), ready (conscientious), and
An alternative approach to search for mediators is to consider the willing (autonomously motivated) if they are to academically succeed.
possible underlying and constituent neural mechanisms that are shared In the present investigation, we examined the additive and interactive
by conscientiousness and autonomous motivation. One possible mech- contributions of these three individual difference variables to the pre-
anism may be indexed by error-related negativity (ERN; Gehring, Goss, diction of undergraduate academic performance. Conscientiousness
Coles, Meyer, & Donchin, 1993), a response-locked event-related poten- contributed more to academic performance at higher levels of intelli-
tial that occurs upon the commission of task-related errors. The ERN is gence, suggesting that an industrious disposition may serve a catalytic
believed to reect the operation of a performance-monitoring system function among students with more intellectual ability. Conscientious-
in the ACC that engages the cognitive-control mechanisms of the ness contributed more to academic performance at lower levels of
prefrontal cortex during response conict when the commission of autonomous motivation, suggesting that an industrious disposition
errors is likely (Yeung et al., 2004). The ERN is a good example of may serve a compensatory function among students with less intrinsic
what a neurophysiological mediator of the interaction effect in question interest. While students may need only to be ready or willing to limit
would resemble for three reasons. First, higher levels of autonomous the probability of poor academic outcomes, students may need to be
motivation predict larger ERN amplitudes (Legault & Inzlicht, 2013). both ready and able if they are to succeed.
Second, larger ERN amplitudes predict higher levels of academic perfor-
mance among university students (Hirsh & Inzlicht, 2010). Finally, Appendix A. Supplementary data
although increasing the motivational importance of a task generally
produces larger ERN amplitudes, conscientiousness has been shown to Supplementary data to this article can be found online at http://
moderate this association, with motivational incentives being effective
only at the lower levels of conscientiousness but not at higher levels
of conscientiousness (Pailing & Segalowitz, 2004). This latter point is
congruent with the current nding in that autonomous motivation
was only predictive of academic performance at lower levels of consci- Aiken, L.S., & West, S.G. (1991). Multiple regression: Testing and interpreting interactions.
entiousness. Future research should thus consider the ERN and related Thousand Oaks, CA: Sage.
Aron, A.R., Robbins, T.W., & Poldrack, R.A. (2004). Inhibition and the right inferior
neurophysiological processes as potential mediators of the presently frontal cortex. Trends in Cognitive Sciences, 8, 170177.
obtained compensatory interaction. 1016/j.tics.2004.02.010.
162 S.I. Di Domenico, M.A. Fournier / Learning and Individual Differences 40 (2015) 156162

Berkman, E.T., Falk, E.B., & Lieberman, M.D. (2011). In the trenches of real-world self- Ning, H.K., & Downing, K. (2012). Inuence of student learning experience on academic
control: Neural correlates of breaking the link between craving and smoking. performance: The mediator and moderator effects of self-regulation and motivation.
Psychological Science, 22, 498506. British Educational Research Journal, 38, 219237.
Black, A.E., & Deci, E.L. (2000). The effects of instructors' autonomy support and students' 01411926.2010.538468.
autonomous motivation on learning organic chemistry: A self-determination theory Noftle, E.E., & Robins, R.W. (2007). Personality predictors of academic outcomes: Big ve
perspective. Science Education, 84, 740756. correlates of GPA and SAT scores. Journal of Personality and Social Psychology, 93,
237X(200011). 116130.
Carroll, J.B. (1993). Human cognitive abilities. New York: Cambridge University Press. Pailing, P.E., & Segalowitz, S.J. (2004). The error-related negativity as a state and trait mea-
Cohen, J., Cohen, P., West, S.G., & Aiken, L.S. (2003). Applied multiple regression/correlation sure: Motivation, personality, and ERPs in response to errors. Psychophysiology, 41,
analysis for the behavioral sciences (3rd ed.). New Jersey, NY: Lawrence Erlbaum 8495.
Associates. Paunonen, S.V., & Ashton, M.C. (2013). On the prediction of academic performance with
Deary, I.J., Penke, L., & Johnson, W. (2010). The neuroscience of human intelligence differ- personality traits: A replication study. Journal of Research in Personality, 47,
ences. Nature Reviews Neuroscience, 11, 201211. 778781.
Deary, I.J., Whalley, L.J., Lemmon, H., Crawford, J.R., & Starr, J.M. (2000). The stability of in- Perry, S.J., Hunter, E.M., Witt, L.A., & Harris, K.J. (2010). P = f (conscientiousness ability):
dividual differences in mental ability from childhood to old age: Follow-up of the Examining the facets of conscientiousness. Human Performance, 23, 342360.
1932 Scottish Mental Survey. Intelligence, 28(1), 4955.
Deci, E.L., & Ryan, R.M. (2000). The what and why of goal pursuits: Human needs Plant, E.A., Ericsson, K.A., Hill, L., & Asberg, K. (2005). Why study time does not predict
and the self-determination of behavior. Psychological Inquiry, 11, 227268. http:// grade point average across college students: Implications of deliberate practice for academic performance. Contemporary Educational Psychology, 30, 96116. http://dx.
Deci, E.L., & Ryan, R.M. (2012). Motivation, personality, and development within embed-
ded social contexts: An overview of self-determination theory. In R.M. Ryan (Ed.), Ox- Richardson, M., Abraham, C., & Bond, R. (2012). Psychological correlates of university
ford handbook of human motivation (pp. 85107). Oxford, UK: Oxford University students' academic performance: A systematic review and meta-analysis. Psychological
Press. Bulletin, 138, 353387.
DeYoung, C.G. (2011). Intelligence and personality. In R.J. Sternberg, & S.B. Kaufman Roberts, B.W., Bogg, T., Walton, K., Chernyshenko, O., & Stark, S. (2004). A lexical
(Eds.), The Cambridge handbook of intelligence (pp. 711737). New York: Cambridge approach to identifying the lower-order structure of conscientiousness.
University Press. Journal of Research in Personality, 38, 164178.
Di Domenico, S.I., Rodrigo, A.H., Ayaz, H., Fournier, M.A., & Ruocco, A.C. (2015). Decision- 6566(03)00065-5.
making conict and the neural efciency hypothesis of intelligence: A functional Roberts, B.W., Caspi, A., & Moft, T.E. (2001). The kids are alright: Growth and stability in
near-infrared spectroscopy investigation. NeuroImage (in press). personality development from adolescence to adulthood. Journal of Personality and
Gehring, W.J., Goss, B., Coles, M.G., Meyer, D.E., & Donchin, E. (1993). A neural system for Social Psychology, 81, 670683.
error detection and compensation. Psychological Science, 4, 385390. http://dx.doi. Roberts, B.W., Chernyshenko, O.S., Stark, S., & Goldberg, L.R. (2005). The structure of con-
org/10.1111/j.1467-9280.1993.tb00586.x. scientiousness: An empirical investigation based on seven major personality ques-
Goldberg, L.R. (1993). The structure of phenotypic personality traits. American Psychologist, tionnaires. Personnel Psychology, 58, 103139.
48, 2634. 6570.2005.00301.x.
Gottfredson, L.S. (1997). Mainstream science on intelligence: An editorial with 52 signa- Roberts, B.W., Jackson, J.J., Fayard, J.V., Edmonds, G.W., & Meints, J. (2009). Conscientious-
tories, history, and bibliography. Intelligence, 24, 1323. ness. In M.R. Leary, & R.H. Hoyle (Eds.), Handbook of individual differences in social be-
S0160-2896(97)90011-8. havior (pp. 369381). New York: Guilford Press.
Hirsh, J.B., & Inzlicht, M. (2010). Error-related negativity predicts academic performance. Roberts, B.W., Walton, K., & Viechtbauer, W. (2006). Patterns of mean-Level change in
Psychophysiology, 47, 192196. personality traits across the life course: A meta-analysis of longitudinal studies.
Jackson, J.J., Wood, D., Bogg, T., Walton, K., Harms, P., & Roberts, B.W. (2010). What do Psychological Bulletin, 132, 125.
conscientious people do? Development and validation of the behavioral indicators Rohde, T.E., & Thompson, L.A. (2007). Predicting academic achievement with cognitive
of conscientiousness (BIC). Journal of Research in Personality, 44, 501511. http://dx. ability. Intelligence, 35, 8392. Ryan, R.M., & Connell, J.P. (1989). Perceived locus of causality and internalization: Exam-
Jang, K.L., Livesley, W.J., & Vemon, P.A. (1996). Heritability of the Big Five personality ining reasons for acting in two domains. Journal of Personality and Social Psychology,
dimensions and their facets: A twin study. Journal of Personality, 64, 577592. 57, 749761. Ryan, R.M., & Deci, E.L. (2008). Self-determination theory and the role of basic
Jang, K.L., McCrae, R.R., Angleitner, A., Riemann, R., & Livesley, W.J. (1998). Heritability psychological needs in personality and the organization of behavior. In O.P.
of facet-level traits in a cross-cultural twin sample: Support for a hierarchical John, R.W. Robins, & L.A. Pervin (Eds.), Handbook of personality: Theory and
model of personality. Journal of Personality and Social Psychology, 74, 15561565. research (pp. 654678) (3rd ed.). New York, NY: Guilford. Sackett, P.R., Gruys, M.L., & Ellingson, J.E. (1998). Abilitypersonality interactions when
Jensen, A.R. (1998). The g factor: The science of mental ability. Westport, CT: Praeger. predicting job performance. Journal of Applied Psychology, 83, 545556. http://dx.
John, O.P., Naumann, L.P., & Soto, C.J. (2008). Paradigm shift to the integrative big-ve tax-
onomy: History, measurement, and conceptual issues. In O.P. John, R.W. Robins, & L.A. Shamosh, N.A., DeYoung, C.G., Green, A.E., Reis, D.L., Johnson, M.R., Conway, A.R.R., et al.
Pervin (Eds.), Handbook of personality: Theory and research (pp. 114158) (3rd ed.). (2008). Individual differences in delay discounting: Relation to intelligence, working
New York, NY: Guilford. memory, and anterior prefrontal cortex. Psychological Science, 19, 904911. http://dx.
Johnson, V.E. (2003). Grade ination: A crisis in college education. New York, NY: Springer-
Verlag. Snijders, T.A.B., & Bosker, R.J. (2012). Multilevel analysis (2nd ed.). An introduction to basic
Komarraju, M., Karau, S.J., & Schmeck, R.R. (2009). Role of the big ve personality traits in and advanced multilevel modeling. London: SAGE.
predicting college students' academic motivation and achievement. Learning and Spearman, C. (1904). General intelligence, objectively determined and measured.
Individual Differences, 19, 4752. American Journal of Psychology, 15(2), 201292.
Kusurkar, R.A., Ten Cate, T.J., Vos, C.M.P., Westers, P., & Croiset, G. (2012). How motivation Spearman, C.S. (1927). The abilities of man. New York: Macmillan.
affects academic performance: A structural equation modelling analysis. Advances in Strenze, T. (2007). Intelligence and socioeconomic success: A meta-analytic review of
Health Sciences Education, 18, 5769. longitudinal research. Intelligence, 35, 401426.
Legault, L., & Inzlicht, M. (2013). Self-determination, self-regulation, and the brain: 2006.09.004.
Autonomy improves performance by enhancing neuroaffective responsiveness to Trautwein, U., Ldtke, O., Roberts, B.W., Schnyder, I., & Niggli, A. (2009). Different forces,
self-regulation failure. Journal of Personality and Social Psychology, 105, 123138. same consequence: Conscientiousness and competence beliefs are independent predictors of academic effort and achievement. Journal of Personality and Social
Loehlin, J.C., McCrae, R.R., Costa, P.T., & John, O.P. (1998). Heritabilities of common and Psychology, 97, 11151128.
measure-specic components of the Big Five personality factors. Journal of Research Vansteenkiste, M., Sierens, E., Soenens, B., Luyckx, K., & Lens, W. (2009). Motivational pro-
in Personality, 32, 431453. les from a self-determination perspective: The quality of motivation matters. Journal
Lubinski, D., Webb, R.M., Morelock, M.J., & Benbow, C.P. (2001). Top 1 in 10,000: A 10- of Educational Psychology, 101, 671688.
year follow-up of the profoundly gifted. Journal of Applied Psychology, 86, 718729. Weinstein, N., & Ryan, R.M. (2010). When helping helps: Autonomous motivation for prosocial behavior and its inuence on well-being for the helper and recipient.
Maier, N.R.E. (1955). Psychology in industry. Boston: Houghton-Mifin. Journal of Personality and Social Psychology, 98, 222244.
McCrae, R.R., & Costa, P.T., Jr. (2008). The ve-factor theory of personality. In O.P. John, a0016984.
R.W. Robins, & L.A. Pervin (Eds.), Handbook of personality: Theory and research Wonderlic, Inc. (2002). Wonderlic Personnel TestTM scholastic level exam user's manual.
(pp. 159181) (3rd ed.). New York, NY: Guilford. Libertyville, IL: Wonderlic.
Miquelon, P., Vallerand, R.J., Grouzet, F.M.E., & Cardinal, G. (2005). Perfectionism, academic Yeung, N., Botvinick, M.M., & Cohen, J.D. (2004). The neural basis of error detection:
motivation, and personal adjustment: An integrative model. Personality and Social Conict monitoring and the error-related negativity. Psychological Review, 111,
Psychology Bulletin, 31, 913924. 931959.
Nakao, T., Mitsumoto, M., Nashiwa, H., Takamura, M., Tokunaga, S., Miyatani, M., et al.
(2010). Self-knowledge reduces conict by biasing one of plural possible answers.
Personality and Social Psychology Bulletin, 36, 455469.