Personality and Individual Differences 131 (2018) 51–56

Contents lists available at ScienceDirect

Personality and Individual Differences

journal homepage: www.elsevier.com/locate/paid

Big Five personality relationships with general intelligence and specific T

Cattell-Horn-Carroll factors of intelligence
⁎
David C. Osmona, , Octavio Santosa, Dmitriy Kazakova, Michelle T. Kassela, Quintino R. Manob,
Ashten Mortha
a
University of Wisconsin – Milwaukee, Department of Psychology, United States
b
University of Cincinnati, Department of Psychology, United States

A R T I C LE I N FO A B S T R A C T

Keywords: Objective: Though relationships among Big Five personality traits and general intelligence have been found
IQ consistently, less is known of how personality traits relate to specific factors of intelligence. The present study
Personality examined relationships between Big Five personality traits on the NEO Personality Inventory-Revised (NEO-PI-R;
Big Five Costa & McCrae, 1992) and g-residualized scores of the seven factors of the Cattell-Horn-Carroll (CHC) model on
IQ factors
the Woodcock-Johnson-III (WJ-III; Mather & Woodcock, 2001).
Method: College students referred to a learning disorders clinic were assessed with a comprehensive neu-
ropsychological battery that involved cognitive, achievement, and personality testing. The sample was culled for
failure on a sensitive effort measure (Word Memory Test; Green, 2003), leaving 140 participants.
Results: Openness (O) accounted for significant variance in all seven WJ-III IQ factors with the O-crystallized
intelligence (Gc) relationship being strongest and all other personality-IQ relationships being small. However,
using residualized scores to remove common intellectual variance showed that Openness related only to Gc while
Extraversion related to both processing speed and Gc.
Conclusions: The role that personality plays in the specific aspects of CHC factors of intellectual test performance
is discussed in the context of learning difficulties in college students.

Substantial relationships between the Big Five factors of personality
and intelligence have been found (e.g., Ashton, Lee, Vernon, & Jang,
2000; Bartels et al., 2012; Bates & Shieles, 2003; Holland, Dollinger,
Holland, & MacDonald, 1995; McRae & John, 1992; Moutafi, Furnham,
& Crump, 2003; von Stumm, Chamorro-Premuzic, Quiroga, & Colom,
2009; Wolf & Ackerman, 2005). Although much of this work has fo-
cused on the relation between personality traits with broad constructs
of intelligence, relatively little is known of how personality traits relate
to more specific factors of intelligence. The purpose of the present study
was to refine our understanding of personality-IQ relationships by ex-
amining the Big Five traits of personality alongside the Cattell-Horn-
Carroll (CHC) seven-factor model. Furthermore, the present study ad-
dressed various criticisms of the extant literature. Specifically, to
identify the relationship of personality traits to factors of intelligence,
those factors must be independent of g, general intelligence factor
(Reeve, Meyer, & Bonaccio, 2006). Additionally, a sample referred for
clinical evaluation of learning difficulties and psychiatric diagnoses was
used that provided a broader range of intellectual performance to avoid
attenuated correlations between personality and intelligence that might
otherwise be due to range restriction in both the personality and in-
telligence measures. Finally, the sample was culled for participants
providing insufficient effort in testing in order to dampen experimental
error associated with mismeasurement of personality and intelligence.
It is still unclear whether personality traits relate just to overall
intelligence or to factors of intelligence, or to both overall and specific
factors of intelligence. As noted by Reeve et al. (2006), various factors
of intelligence have shown different relationships with the Big Five
traits across studies (e.g., -N: Austin, Hofer, Deary, & Eber, 2000; +E:
Wolf & Ackerman, 2005; A: Krebs & Sturrup, 1982; –C: Furnham &
Chamorro-Premuzic, 2005). Nevertheless, it might be argued that the
most consistent personality-IQ relationships holds for Openness
(Moutafi et al., 2003; Zeidner & Matthews, 2000). Specifically, Ashton
et al. (2000) found Openness to correlate better with crystallized in-
telligence (Gc) than fluid intelligence (Gf), which is consistent with
previous studies (e.g., Ackerman & Goff, 1994; Goff & Ackerman,
1992). Similarly, a meta-analysis confirmed the relationship of Open-
ness to Gc over other factors of intelligence (Ackerman & Heggestad,
1997, although see Reeve et al., 2006 and von Stumm et al., 2009 for a

⁎
Corresponding author at: University of Wisconsin-Milwaukee, Department of Psychology, 2441 E. Hartford Ave., Milwaukee, WI 53211, United States.
E-mail address: neuropsy@uwm.edu (D.C. Osmon).

https://doi.org/10.1016/j.paid.2018.04.019
Received 14 December 2017; Received in revised form 6 April 2018; Accepted 11 April 2018
0191-8869/ © 2018 Elsevier Ltd. All rights reserved.
D.C. Osmon et al.
divergent opinion). Despite the weight of evidence favoring the Open-
ness-Gc relationship, criticisms of existing evidence need to be ad-
dressed. First, a systematic analysis using a CHC model of intelligence is
lacking in present literature, and it can therefore not be ruled out that
additional Big Five personality traits may differentially relate to the
various factors of intelligence. Such a model (crystallized intelligence
[Gc], long-term retrieval [Glr], visual-spatial thinking [Gv], auditory
processing [Ga], short-term memory [Gsm], fluid reasoning [Gf], and
processing speed [Gs]) represents the synthesis of a comprehensive
account of intelligence and, when measured in a single battery, can
provide a more comprehensive study of the personality-intelligence
relationship. Second, consistent with Reeve et al. (2006) specific factors
of intelligence that are not substantially correlated with g must be ex-
amined for their association to personality variables. Only in this way
can personality be related to specific intellectual variance apart from
overall intelligence.
Another confounding variable that may be clouding the in-
telligence-personality relationship is insufficient effort in test perfor-
mance. Poor effort is a nuisance variable that confounds assessment
literature results (Green, 2007). Thus, ensuring adequate effort in the
participant sample can lessen experimental error, yielding a better ex-
amination of the personality relationship with intelligence. For ex-
ample, college student samples show a 5–12% failure rate on sensitive
measures of effort (Ross et al., 2015; Santos, Kazakov, Reamer, Park, &
Osmon, 2014; Silk-Eglit et al., 2014). Likewise, poor participant effort
and/or feigning/malingering have been found to explain sizeable
amounts of variance in neuropsychological performance, especially
when incentive to falsify performance is present (e.g., to obtain At-
tention-Deficit/Hyperactivity Disorder (ADHD) medication;
Constantinou, Bauer, Ashendorf, Fisher, & McCaffrey, 2005; Green,
2007; Green, Rohling, Lees-Haley, & Allen, 2001; Williamson et al.,
2004). Furthermore, poor participant effort has been found to operate
even when no discernable incentive to appear deficient has been found.
For example, An, Zakzanis, and Joordens (2012) found that almost half
of college participants in a study of neuropsychological test perfor-
mance failed a common performance validity measure. While this
finding is probably not a good estimate of college student effort in
general (see Ross et al., 2015; Santos et al., 2014; Silk-Eglit et al.,
2014), it does suggest that under some circumstances effort problems
can grossly affect experimental findings. Therefore, well-designed as-
sessment studies must ensure internal validity by at least excluding
participants for failure on a sensitive measure of performance validity,
such as Green's Word Memory Test (WMT) or Medical Symptom Va-
lidity Test (MSVT).
The present study used a sample of participants that excluded for
insufficient effort in order to minimize experimental error in evaluating
the relationship between personality and overall intelligence and its
various components. Because there may be a differential relationship
between each personality trait and the different factors of intelligence,
the Woodcock-Johnson-III (WJ-III; Mather & Woodcock, 2001) instru-
ment was used to assess seven CHC factors of intelligence as well as a
full-scale score (GIA). While the WJ-III provides an aggregate in-
telligence score (GIA) as well as specific intelligence scores (Gc, Gf, Glr,
Gsm, Ga, Gv, Gs; see Materials below for description), each of these
specific scores correlate substantially with the aggregate score and,
therefore, do not purely represent specific intelligence factors. In order
to mitigate the concern about the overestimation of personality re-
lationships with specific factors of intelligence (see Reeve et al., 2006,
for a discussion), residualized scores for the CHC factor scores were
obtained by removing common intelligence variance (GIA) and then
were subjected to additional regression analyses for comparison to non-
residualized CHC factor scores.
52
Personality and Individual Differences 131 (2018) 51–56
1. Methods
1.1. Participants
Participants initially included a total of 240 individuals referred to
the university Learning Disability Clinic for academic concerns. Sixty-
two of those participants were excluded for failing the Word Memory
Test (any of the IR, DR, CNS, MC, PA scales (Green, 2003); see the
following paragraph for details) leaving 178 participants of which 145
had an available primary diagnosis/no diagnosis: ADHD (38), Psy-
chiatric (28), Cognitive Disorder-NOS (10), Learning Disorder (LD)-
NOS (8), Math LD (8), No Warranted Diagnosis (17), Reading LD (35),
Writing LD (1). Thirty-six percent of the sample had co-morbid diag-
noses and 28 of the participants were found to have only psychiatric
diagnoses but were included to fairly represent a sample referred for
academic problems. The sample had an average age of 26.71
(SD = 9.66) and were college students who were given a multitude of
psychoeducational measures for clinical purposes, which included the
Woodcock-Johnson cognitive and achievement tests (WJ-III), stand-
alone specific reading measures (e.g., Test of Word Reading Efficiency),
neuropsychological measures (e.g., memory, attention, executive
function tests), broad-based personality (i.e., NEO-PI-R) and college
adjustment measures (e.g., Achievement Motivation Profile, Career
Decisions Inventory, SACQ, etc.). Because this battery required on
average 11.5 h to complete, participants were tested over two different
sessions on consecutive days. Participants who were fluent in English
and who completed the majority of measures including the complete
WJ-III cognitive and NEO-PI-R were included in the analyses. Five
participants had incomplete data on the WJ-III cognitive and/or NEO-
PI-R, thus all analyses were conducted on the remaining 140 partici-
pants.
Sixty-two (26%) of the 240 total participants were removed from
final analyses due to inadequate effort defined by performance cutoff
scores below threshold for any of the five symptom validity scales of the
Word Memory Test (IR, DR, CNS, MC, PA (Green, 2003)). Fig. 1 shows
the entire distribution of participants on the WJ-III GIA compared to the
distribution with participants removed for poor effort. Both distribu-
tions were fitted adequately by a normal distribution with delta Akai-
ke's Information Criterion-corrected (AICc) values within < 2 of the
best fitting distributions (Johnson Su and Gamma, respectively). Like-
wise, distributions for the two NEO-PI-R variables were also modeled by
multiple distributions. Openness fit a normal distribution both before
and after culling participants. Conscientiousness fit a Normal 2 Mixture
distribution according to delta AICc values. Nevertheless, the distribu-
tion was normal looking except for a slight overrepresentation at the
low end of the distribution. Additionally, removing participants for
poor effort nearly eliminated outliers in the GIA distribution, as evident
in the box plot. Since culling participants improved the distribution of
the retained sample, further analyses were completed with the retained
sample. Additionally, parametric statistics appear to be warranted for
these distributions, even for Conscientiousness because the statistics are
robust in near-normal distributions.
1.2. Materials
The NEO Personality Inventory-Revised (NEO-PI-R; Costa &
McCrae, 1992) is a 240-item questionnaire that measures the Big Five
personality traits: Neuroticism, Extraversion, Openness to Experience,
Agreeableness, and Conscientiousness, each of which is comprised of six
facets. NEO-PI-R measures used in the study included composites (mean
of 50; standard deviation of 10) of the Big Five variables. Neuroticism
(N) is the predisposition to experience psychological stress as mani-
fested by anxiety, anger, depression or other negative affect. Ex-
troversion (E) includes sociability, liveliness, and cheerfulness. Open-
ness (O) comprises aesthetic sensitivity, intellectual curiosity, need for
variety, and non-dogmatic attitudes. Agreeableness (A) consists of trust,
D.C. Osmon et al. Personality and Individual Differences 131 (2018) 51–56

Fig. 1. WJ-III GIA and NEO-PI-R Openness and Conscientiousness distributions highlighting the participants removed from the sample for failing at least one of the
five symptom validity scales on the Word Memory Test (IR, DR, CNS, MC, PA; excluded participants are displayed in dark color on top row). Bottom row shows the
distributions with only the participants retained in the sample (N = 140), noting that normal distributions in all three cases are adequate fits indistinguishable (AIC-
c < 2) from the best fitting distributions.

altruism, and sympathy. Conscientiousness (C) encompasses a dis-
ciplined striving after goals and a strict adherence to principles. Com-
posites from these five personality traits were used in the study.
The Woodcock-Johnson III (WJ-III; Mather & Woodcock, 2001)
Tests of Cognitive Abilities comprise a wide age-range and compre-
hensive system for measuring general and specific intellectual abilities.
WJ-III measures used in the study included standard scores (mean of
100; standard deviation of 15) for the General Ability Index (GAI:
overall intelligence quotient), Gc (crystallized intelligence composite),
Gf (fluid intelligence composite), Glr (long-term retrieval composite),
Gsm (short-term memory composite), Gv (visual processing composite),
Ga (auditory processing composite), and Gs (processing speed compo-
site).
1.3. Procedure
All of the testing was done within the university Learning Disability
Clinic. Prior to the evaluation, informed consent was obtained from
each participant subsequent to outlining the purpose, procedure, con-
fidentiality, possible risks and benefits of the evaluation, and informing
participants that they could withdraw from the evaluation at any time.
Each participant was tested within the Learning Disability Clinic across
two different sessions on two consecutive days totaling about 11 to
12 h. The participants and their data were treated in accordance with
Institutional Review Board procedures.
1.4. Analyses
All statistics were completed using JMP Pro 12 (SAS, 2015). Dis-
tributional analyses were run to ensure appropriateness of distribu-
tional statistics and to determine whether excluding participants on the
basis of effort measure failure was necessary. Akaike's Information
Criterion-corrected (AICc) values were used to examine the best fitting
distribution and box plots were employed to detect outliers since these
values can have serious consequences in correlational analyses.
Forward regression analyses were conducted to examine how well
the NEO-PI-R predicted the GIA and 7 CHC intelligence factors both
separately and in combination. A total of 8 hierarchical regressions
were run using the 5 NEO-PI-R variables to predict separately the GIA
and the 7 intellectual factors, with Openness forced into the equation
first because of its clear high correlation with all the intellectual factors
relative to the other 4 Big Five variables. The minimum Bayesian

53
D.C. Osmon et al. Personality and Individual Differences 131 (2018) 51–56

Information Criterion (BIC) was used to select the best model fit be-
cause of its greater penalty of models with the greater number of
parameters compared to the Akaike Information Criterion (AIC).
In order to evaluate the concern that common variance among
factors of intelligence leads to overestimation of the role of personality
(Reeve et al., 2006), another seven hierarchical regressions were run on
the residual scores of each of the seven specific factor IQ scores (Gc, Gf,
Glr, Gsm, Ga, Gv, Gs). Residual specific IQ scores were obtained by
regressing each factor on overall intelligence (GIA). This procedure
removed common IQ variance, yielding residual scores that reflected
specific IQ factor scores. Those residual scores were then predicted by
each of the Big Five personality variables, forcing Openness first into
the equation.
All 16 regression analyses used K-fold cross validation procedures to
correct for overfitting a model of personality-IQ relationships. K-fold
cross validation uses subsets of data to validate the model (five subsets
were used in the current analyses), yielding a K-fold R2 value that is the
average of the K-fold R2 values of the subsets and represents a better
estimate of the predictive value of the model. Both the R2adj and the K-
fold R2 are reported to give a sense of the over-fitting of the non-vali-
dated model.
2. Results
2.1. IQ-NEO relationships
Variance was predicted in all of the WJ-III GIA and 7 CHC factors by
the NEO-PI-R Openness score (see Table 1; values range from F
[1137] = 7–48, p < .001) according to hierarchical regression with
order determined by Openness being forced first into the regression
equation and the other four personality traits loading according to
probability value (p < .05). Moreover, Conscientiousness, and Agree-
ableness scores predicted small amounts of variance in addition to that
predicted by Openness in the GIA and Gs scores, respectively (values
range from F[1137] = 5–7, p < .03), while Neuroticism predicted
additional variance beyond Openness in Gsm (values range from F
[1137] = 4, p < .05). Multiple correlations (see Table 1) were positive
indicating that higher intellectual ability was associated with greater
levels of Openness and Neuroticism, while negative correlations in-
dicated that higher intellectual function was associated with lower
Agreeableness for Gs, and lower Conscientious for GIA. No other NEO-
PI-R variables predicted significant variance in any of the IQ variables.
Table 1 also shows the percent variance accounted for when corrected
for attenuation due to unreliability of measures (R2~) and range re-
striction (R2’) using values of reliability from Schrank, McGrew, and
Woodcock (2001) and McCrea, Kurtz, Yamagata, and Terracciano
(2011).
Using residualized specific IQ factor scores yielded different results,
reducing the relationship between personality and specific IQ scores.
Openness and extraversion each predicted about 3% of the variance in
Gc, respectively (F[2134] = 4.87, p < .01, R2adj = 0.05; K-
foldR2 = 0.005) with Openness showing a positive relationship to Gc
while Extraversion showed a negative relationship to Gc. The only other
significant result was for Gs (F[2139] = 4.95, p < .01). Here, only
Extraversion contributed to the prediction (R2adj = 0.05; K-
foldR2 = 0.04), showing a positive relationship to Gs.
3. Discussion
The current results are consistent with the bulk of studies that have

Table 1
Percent variance accounted for in WJ-III variables by NEO-PI and increases in variance accounted for when correlations are corrected for attenuation due to
unreliability of measures (rel-R2) and range restriction (rr-R2).
WJ-III Openness Conscientiousness Extraversion Agreeableness Neuroticism

rr-R2/rel-R2 rr-R2/rel-R2 rr-R2/rel-R2 rr-R2/rel-R2 rr-R2/rel-R2

GIA .26** −.02* – – –

0.37/0.30 0.02/0.03 –/– –/– –/–
Gc .28** 0.01 0.01 – –
0.41/0.34 –/– –/– –/– –/–
Gc-r .03* – −.03* – –
0.04/0.03 –/– −0.04/−0.04 –/– –/–
Glr .13** – – – –
0.18/0.17 –/– –/– –/– –/–
Glr-r – – – – –
–/– –/– –/– –/– –/–
Gv .07** – – – –
0.15/0.11 –/– –/– –/– –/–
Gv-r – – – – –
–/– –/– –/– –/– –/–
Ga .25** 0.01 – – –
0.54/0.33 –/– –/– –/– –/–
Ga-r 0.02 – – – –
0.02/0.03 –/– –/– –/– –/–
Gf .14** – – – –
0.22/0.22 –/– –/– –/– –/–
Gf-r – – – – 0.01
–/– –/– –/– –/– –/–
Gs .04* – – −.04* –
0.05/0.06 –/– –/– −5/−5 –/–
Gs-r 0.02 – .05** – –
0.02/0.02 –/– 0.05/0.06 –/– –/–
Gsm 0.14 – – – 0.03
0.15/0.15 –/– –/– –/– 0.04/0.04
Gsm-r – – – – –
–/– –/– –/– –/– –/–

Note. WJ-III = Woodcock Johnson-3rd edition; GIA = General intellectual ability; Gc = crystallized; Glr = long-term retrieval; Gv = visual; Ga = auditory;
Gf = fluid reasoning; Gs = processing speed; Gsm = short-term memory; xx-r = residual score version; Minus sign means an inverse relationship.* p < .05; **
p < .01; rr-R2 = correlation corrected for range restriction, rel-R2 = correlation corrected for attenuation due to unreliability; – = < 1%.

54
D.C. Osmon et al.
found the strongest personality-IQ correlation between Openness and
Gc (Ackerman & Heggestad, 1997; Ashton et al., 2000; Goff &
Ackerman, 1992; McCrae & Costa, 1985). They are also consistent with
Ackerman and Heggestad (1997) who found school performance and
the knowledge aspect of Gc to be related to Conscientiousness and
Openness (and especially its cognate: Typical Intellectual Engagement,
or amount of intellectual effort “typically” proffered in a work situa-
tion). The current results replicated the relationship between Openness
and various intellectual factors but did not show Conscientiousness to
be related to any of the CHC seven factors. Past work along with the
current results reinforces the notion that Conscientiousness relates to
academic and general life success irrespective of intellectual ability
(Ackerman & Goff, 1994; Chamorro-Premuzic & Furnham, 2008; Noftle
& Robins, 2007; Paunonen & Ashton, 2001). Below is a brief discussion
of results.
While the current results were generally consistent with past re-
search, the fact that we did not find Neuroticism to be strongly asso-
ciated with any CHC factors (except the 3% variance explained in Gsm)
stands in contrast to prior results. For example, Ackerman and
Heggestad (1997) noted a consistent finding that Neuroticism was ne-
gatively correlated with intelligence while a positive relationship to
intelligence was found in the present results. This discrepancy in the
direction of the Neuroticism correlation may be related to measurement
issues. That is, Neuroticism as measured by the NEO-PI-R may involve
fewer psychopathological aspects and be more related to stress man-
agement/tolerance. As a result, Neuroticism, as measured in a more
psychopathological fashion, might carry the negative trait relationship
with intelligence (à la Moutafi et al., 2003) while Neuroticism as
measured by the NEO-PI-R might carry the positive trait relationship to
intelligence.
Methodological issues have been noted to cloud personality-IQ re-
lationships (Reeve et al., 2006; von Stumm et al., 2009), especially
those that relate to confounding common and specific intellectual
variances. These concerns were addressed in current results by using
the WJ-III residual scores of the specific factors (Gc, Gf, Glr, Gsm, Ga,
Gv, Gs), derived by removing common intelligence variance (GIA).
Doing so supported the concerns of Reeve et al. (2006) that common
intellectual variance both confounded and inflated the relationship
between personality and specific factors of intelligence. The current
results indicate that personality scores (specifically Openness and
Conscientiousness) accounted for about 28% of the variance in overall
intelligence. When general intellectual variance was removed by using
residualized intellectual factor scores, personality relationships were
much reduced and smaller, accounting for only 2–3% of variance.
Specifically, only residualized Openness and residualized Extraversion
accounted for significant variance in Gc and only residualized Extra-
version related to Gs. No other residualized factor scores had relation-
ships to personality.
The present results clarify some inconsistencies in the literature by
suggesting that Openness is mainly related to general intellectual power
(g), and that some specific cognitive abilities are weakly related to
Extraversion, Conscientiousness, and Neuroticism. Specifically, being
more open to experience does relate to the knowledge acquisition as-
pects of intelligence, consistent with the speculation that Openness
relates to “interests in knowledge” or Gc (e.g., Ackerman & Heggestad,
1997; Bates & Shieles, 2003; Cattell, 1943; Chamorro-Premuzic &
Furnham, 2006; Moutafi, Furnham, & Paltiel, 2005). Interestingly, the
current results also suggest that being less Extraverted is associated
with Gc independent of general intellectual power. Unfortunately, the
Gc results may not be reliable given the low K-fold R2 value compared
to the R2adj value. This weak result was surprising given the balance of
results in the literature suggesting the strength of the Openness-Gc re-
lationship. Finally, the current results suggest that being more Extra-
verted is associated with greater mental speed independent of general
intellectual power. Notably, this result may be more reliable given the
small drop off in K-fold R2 value compared to the R2adj value.
55
Personality and Individual Differences 131 (2018) 51–56
The current results do not support the past findings that greater
intelligence is associated with positive personality traits because one
positive trait, specifically Extraversion, showed a negative relationship
to Gc while another positive trait, specifically Openness, showed a
positive relationship to both GIA and Gc. Likewise, the negative per-
sonality trait (Neuroticism) did not relate to any residualized specific
intellectual factor or to GIA.
Limitations of the present results relate largely to the particular
sample used and to issues associated with the Openness construct, as
noted in three following points. First, the present results are general-
izable only to college students seeking evaluation for learning difficulty,
although meta-analysis results have provided preliminary support for
the O-Gc relationship being more widespread across the population at
large and across a greater age range (Ackerman & Heggestad, 1997).
The relatively large number of participants and the large effect sizes for
the relationship of Openness with GIA suggest that these results are
robust, as does the generally consistent findings with past literature.
However, the smaller relationships of personality to residualized in-
tellectual factors suggest that personality has a limited effect in specific
intellectual factors. Additionally, the O-Gc finding is likely to be un-
reliable, although the E-Gs result may be more robust in cross valida-
tion. Future studies with a larger sample size are desirable and using
residualized scores for intellectual factors will be important in testing
how robust the current findings are in cross-validation. Second, some
have criticized the Openness trait on theoretical grounds arguing that it
is “the most debatable of the five dimensions” that lacks cohesiveness
and replicability with at least three different interpretations of “Cul-
ture…Openness…and Intellect” (Digman, 1990, pp. 433). Third, accu-
rate relationships between two constructs that are measured with dif-
fering methodologies (i.e., questionnaire and ability tests) may be
called into question. For example, as noted by one anonymous re-
viewer, self-estimates of intelligence may show stronger correlations
with some personality questionnaire-based constructs than actual
measured intelligence.
Despite these limitations, the present results appear to offer robust
findings regarding relatively strong relationships between personality
and general intellectual power. It is unclear why the present results
yielded higher correlations than did previous studies (r ~ 0.4–0.5 vs.
r ~ 0.3–0.4 in other studies), although excluding participants for subpar
effort, using an IQ measure built upon the theoretical foundation of
CHC theory, and using a population of participants with impairments
that increase variance in cognitive functioning may help explain the
stronger results. Such strong relationships need greater consideration in
clinical work with college students with learning difficulties because
personality traits may be protective for commonly lower intellectual
ability. There are also implications of the present results for intellectual
and neuropsychological assessment in general. The consistency of per-
sonality-IQ relationships in the literature and validation of this litera-
ture in the current results requires clinicians to consider the importance
of openness when interpreting the meaning of intellectual performance.
In clinical psychology, there is a tendency to consider personality and
intelligence as independent constructs but that is incommensurate with
the 25–30% of shared variance in intellectual performance and the
openness and conscientiousness personality traits.
Finally, the present findings warrant exploration of the underlying
cause of the correlation of Openness with general intellectual power,
and the small but apparently robust relationship between Extraversion
and mental speed. While it is difficult to envision a true experimental
paradigm for examining causative relationships between personality
and intelligence, it is conceivable that “openness” induction procedures
could be used to investigate intellectual performance. For example,
viewing films of open versus closed individual relations preceding
problem-solving tasks may be one way to attempt to induce openness
attitudes and assess subsequent changes in intellectual performance.
Additionally, more quasi-experimental paradigms of the sort used by
Gregory, Nettelbeck, and Wilson (2010) would also be useful, although
D.C. Osmon et al. Personality and Individual Differences 131 (2018) 51–56

using less “Gc-saturated” participants may be a better methodology for Gregory, T., Nettelbeck, T., & Wilson, C. (2010). Openness to experience, intelligence, and
this purpose. successful ageing. Personality and Individual Differences, 48(8), 895–899.
Holland, D. C., Dollinger, S. J., Holland, C. J., & MacDonald, D. A. (1995). The re-
lationship between psychometric intelligence and the five-factor model of personality
References in a rehabilitation sample. Journal of Clinical Psychology, 51, 79–88.
Krebs, D., & Sturrup, B. (1982). Role-taking ability and altruistic behaviour in elementary
school children. Journal of Moral Education, 11(2), 94.
Ackerman, P. L., & Goff, M. (1994). Typical intellectual engagement and personality:
Mather, N., & Woodcock, R. W. (2001). Examiner's manual: Woodcock-Johnson III tests of
Reply to Rocklin (1994). Journal of Educational Psychology, 86, 150–153.
cognitive abilities. Itasca, IL: Riverside Publishing.
Ackerman, P. L., & Heggestad, E. D. (1997). Intelligence, personality, and interests:
McCrae, R. R., & Costa, P. T. (1985). Updating Norman's “adequate taxonomy”:
Evidence for overlapping traits. Psychological Bulletin, 121(2), 219–245.
Intelligence and personality dimensions in natural languages and in questionnaires.
An, K. Y., Zakzanis, K. K., & Joordens, S. (2012). Conducting research with non-clinical
Journal of Personality and Social Psychology, 49, 710–721.
healthy undergraduates: Does effort play a role in neuropsychological test perfor-
McCrea, R. R., Kurtz, J. E., Yamagata, S., & Terracciano, A. (2011). Internal consistency,
mance? Archives of Clinical Neuropsychology, 27(8), 849–857.
retest reliability, and their implications for personality scale validation. Personality
Ashton, M. C., Lee, K., Vernon, P. A., & Jang, K. L. (2000). Fluid intelligence, crystallized
and Social Psychology Review, 15, 28–50.
intelligence, and the openness/intellect factor. Journal of Research in Personality, 34,
McRae, R. R., & John, O. P. (1992). An introduction to the five-factor model and its
198–207.
applications. Journal of Personality, 60(2), 175–215.
Austin, E. J., Hofer, S. M., Deary, I. J., & Eber, H. W. (2000). Interactions between in-
Moutafi, J., Furnham, A., & Crump, J. (2003). Demographic and personality predictors of
telligence and personality: Results from two large samples. Personality and Individual
intelligence: A study using the NEO-personality inventory and the Myers–Briggs type
Differences, 29(3), 405–427.
indicator. European Journal of Personality, 17, 79–94.
Bartels, M., van Weegen, F. I., van Beijsterveldt, C. E. M., Carlier, M., Polderman, T. J. C.,
Moutafi, J., Furnham, A., & Paltiel, L. (2005). Can personality factors predict intelligence?
Hoekstra, R. A., & Boomsma, D. I. (2012). The five factor model of personality and
Personality and Individual Differences, 38, 1021–1033.
intelligence: A twin study on the relationship between the two constructs. Personality
Noftle, E. E., & Robins, R. W. (2007). Personality predictors of academic outcomes: Big
and Individual Differences, 53, 368–373.
Five correlates of GPA and SAT scores. Journal of Personality and Social Psychology, 93,
Bates, T. C., & Shieles, A. (2003). Crystallized intelligence as a product of speed and drive
116–130.
for experience: The relationship of inspection time and openness to g and Gc.
Paunonen, S. V., & Ashton, M. C. (2001). Big Five predictors of academic achievement.
Intelligence, 31, 275–287.
Journal of Research in Personality, 35, 78–90.
Cattell, R. B. (1943). The description of personality: Basic traits resolved into clusters.
Reeve, C. L., Meyer, R. D., & Bonaccio, S. (2006). Intelligence–personality associations
Journal of Abnormal and Social Psychology, 38(4), 476–506.
reconsidered: The importance of distinguishing between general and narrow di-
Chamorro-Premuzic, T., & Furnham, A. (2006). Intellectual competence and the in-
mensions of the intelligence. Intelligence, 34, 387–402. http://dx.doi.org/10.1016/j.
telligent personality: A third way in differential psychology. Review of General
intell.2005.11.001.
Psychology, 10, 251−267.
Ross, T. P., Poston, A. M., Rein, P. A., Salvatore, A. N., Wills, N. L., & York, T. M. (2015).
Chamorro-Premuzic, T., & Furnham, A. (2008). Personality, intelligence, and approaches
Performance invalidity base rates among healthy undergraduate research partici-
to learning as predictors of academic performance. Personality and Individual
pants. Archives of Clinical Neuropsychology, 31, 97–104.
Differences, 44(7), 1596–1603.
Santos, O. A., Kazakov, D., Reamer, M. K., Park, S. E., & Osmon, D. C. (2014). Effort in
Constantinou, M., Bauer, L., Ashendorf, L., Fisher, J., & McCaffrey, R. J. (2005). Is poor
college undergraduates is sufficient on the Word Memory Test. Archives of Clinical
performance on recognition memory effort measures indicative of generalized poor
Neuropsychology, 29(7), 609–613.
performance on neuropsychological tasks? Archives of Clinical Neuropsychology, 20(2),
SAS (2015). JMP 12 basic analysis. Cary, NC: SAS Institute, Inc.
191–198.
Schrank, F. A., McGrew, K. S., & Woodcock, R. W. (2001). Technical abstract (Woodcock-
Costa, P. T., & McCrae, R. R. (1992). Professional manual: Revised NEO personality inventory
Johnson III Assessment Service Bulletin No. 2). Itasca, IL: Riverside Publishing.
and NEO five-factor inventory. Lutz, FL: Psychological Assessment Resources, Inc.
Silk-Eglit, G. M., Stenclik, J. H., Gavett, B. E., Adams, J. W., Lynch, J. K., & McCaffrey, R.
Digman, J. M. (1990). Personality structure: Emergence of the five-factor model. Annual
J. (2014). Base rate of performance invalidity among non-clinical undergraduate
Review of Psychology, 41, 417–440.
research participants. Archives of Clinical Neuropsychology, 29(5), 415–421.
Furnham, A., & Chamorro-Premuzic, T. (2005). Personality and intelligence: Gender, the
von Stumm, S., Chamorro-Premuzic, T., Quiroga, M. A., & Colom, R. (2009). Separating
big five, self-estimated and psychometric intelligence. International Journal of
narrow and general variances in intelligence-personality associations. Personality and
Selection and Assessment, 13(1), 11–24.
Individual Differences, 47, 336–341.
Goff, M., & Ackerman, P. L. (1992). Personality–intelligence relations: Assessment of
Williamson, D. J., Drane, D. L., Stroup, E. S., Miller, J. W., Holmes, M. D., & Wilensky, A.
typical intellectual engagement. Journal of Educational Psychology, 84, 537–553.
J. (2004). Detecting cognitive differences between patients with epilepsy and patients
Green, P. (2003). Word memory test for Windows: User's manual. Edmonton, Alberta:
with psychogenic nonepileptic seizures: Effort matters. Epilepsia, 45(Suppl. 7), 179.
Green's Publishing.
Wolf, M. B., & Ackerman, P. L. (2005). Extraversion and intelligence: A meta-analytic
Green, P. (2007). The pervasive influence of effort on neuropsychological tests. Physical
investigation. Personality and Individual Differences, 39, 531–542.
Medicine and Rehabilitation Clinics of North America, 18(1), 43–68.
Zeidner, M., & Matthews, G. (2000). Intelligence and personality. In R. Sternberg (Ed.).
Green, P., Rohling, M., Lees-Haley, P., & Allen, M. (2001). Effort has a greater effect on
Handbook of intelligence (pp. 581–610). New York: Cambridge University Press.
test scores than severe brain injury in compensation claimants. Brain Injury, 15(12),
1045–1060.

56