PERSONALITY PROCESSES AND INDIVIDUAL

DIFFERENCES

Functional and Dysfunctional Impulsivity:

Personality and Cognitive Correlates
Scott J. Dickman
University of Texas at Austin

The 3 studies reported here were designed to clarify the nature of the personality trait of impulsivity.
Two types of impulsivity were distinguished. Dysfunctional impulsivity is the tendency to act with
less forethought than most people of equal ability when this tendency is a source of difficulty; most
previous work on impulsivity appears to have focused on this trait. Functional impulsivity, in con-
trast, is the tendency to act with relatively little forethought when such a style is optimal. The present
work indicates that these two tendencies are not highly correlated and that they bear different re-
lations both to other personality traits and to the manner in which certain basic cognitive processes
are executed.

A substantial body of evidence suggests that individual
differences in the personality trait of extraversion are associated
with individual differences in the way basic cognitive processes
are carried out. These basic cognitive processes include the re-
trieval of information from short- and long-term memory
(Bone, 1971; M. W. Eysenck, 1974; McLaughlin, 1968), signal
detection (Harkins & Geen, 1975), probability learning (Wal-
lach & Gahm, 1960), and vigilance (Bakan, 1959).
Factor analyses of extraversion scales indicate that this broad
trait consists of several closely associated traits. These include
impulsivity, sociability, adventurousness, enthusiasm, high ac-
tivity level, and boredom proneness (S. B. G. Eysenck & Eys-
enck, 1963, 1977, 1978; Gerbing, Ahadi, & Patton, 1987; J. P.
Guilford, 1975; Howarth& Browne, 1972;Plomin, 1976).
Recently, researchers have attempted to identify the specific
characteristics of extroverted individuals that account for the
overall relation between extraversion and cognitive functioning.
This research suggests that impulsivity is the component of ex-
traversion that is most consistently associated with individual
differences in the way very basic perceptual and memorial pro-
cesses are executed (e.g., Anderson & Revelle, 1983; Dickman,
1985; Dickman & Meyer, 1988; H.J. Eysenck & Eysenck, 1985;
H. J. Eysenck & Levey, 1972; Loo, 1979; Revelle, Humphreys,
Simon, & Gilliland, 1980).
The present work was designed to examine in more detail
this personality-cognition relation. The studies reported here
sought to distinguish between different types of impulsivity and
to determine which type accounts for the relationship found
between broad measures of impulsivity and measures of cogni-
functional impulsivity, and the latter trait will be referred to as
tive functioning.
Impulsivity can be denned as the tendency to deliberate less
than most people of equal ability before taking action. The con-
sequences of this lack of deliberation for cognitive functioning
seem to be viewed generally as negative. For example, there is a
large body of literature on techniques for reducing impulsivity
(e.g., Cole & Hartley, 1978; Denny, 1970;Heider, 1971;Kagan,
Pearson, & Welch, 1966; Meichenbaum & Goodman, 1971;
Nelson & Birkimer, 1978; Ridberg, Parke, & Hetherington,
1971; Zelniker, Jeffrey, Ault, & Parsons, 1972); there does not
appear to be any corresponding body of work on training low
impulsives to be more impulsive.
Recent work on the relationship between impulsivity and
cognitive functioning, however, suggests that the consequences
of impulsivity are not always negative. For example, when the
experimental task is very simple, high impulsives' rapid re-
sponding has little cost in errors (Dickman, 1985). And when
the time available for making a decision is extremely brief, high
impulsives are actually more accurate than low impulsives
(Dickman & Meyer, 1988).
One question raised by thesefindingsis whether the factors
that cause people to respond quickly and inaccurately when this
style of responding is a source of difficulty are the same factors
that cause people to respond quickly and inaccurately when this
style is optimal. It could be that there is a general tendency to
respond quickly and inaccurately that is sometimes a source of
difficulty and sometimes beneficial. Or, it could be that there
are two separate traits, one that results in rapid inaccurate per-
formance in situations where this is optimal and the other that
results in rapid, inaccurate performance in situations where
this is nonoptimal. The former trait will be referred to here as
dysfunctional impulsivity. The studies described here sought to
determine whether these two traits can be empirically distin-
guished.

Correspondence concerning this article should be addressed to Scott Study 1

J. Dickman, Department of Psychology, University of Texas, Austin, The aim of thefirststudy was to determine whether it is possi-
Texas 78712. ble to discriminate within the self-report domain between func-
Journal of Personality and Social Psychology, 1990, Vol. 58, No. 1,95-102
Copyright 1990 by the American Psychological Association, Inc. 0022-3514/90/$00.7S

95
96 SCOTT J. DICKMAN
Table 1
Eigenvalues and Percentage of Variance Accountedfor
by the Factors in Study 1
Variance
Factor Eigenvalue accounted for (%)
1 8.8 23.0
2 6.0 15.0
3 2.8 7.2
4 1.7 4.2
5 1.2 3.0
6 1.1 2.7
7 1.0 2.6
tional and dysfunctional impulsivity. The approach taken was
to write a large number of items designed specifically to tap
these two types of impulsivity and then to factor-analyze sub-
jects' responses to these items in order to see whether the two
hypothesized factors would emerge.
Method
Measures
The item pool used in this factor analysis contained 17 items written
to tap functional impulsivity and 23 items designed to tap dysfunctional
impulsivity. There were also 23filleritems.
Procedure
The items were administered to 477 undergraduates (209 men and
268 women) who were serving in the study to fulfill a course require-
ment. Subjects were tested in groups of 10-20 persons. Each session
lasted approximately half an hour.
Results
The correlation matrix for the 63 items was factor-analyzed
using the principal axis method, with the communalities esti-
mated iteratively. The criterion for factor extraction was an ei-
genvalue equal to or greater than 1. Seven factors met this crite-
rion. The eigenvalues and the percentage of variance accounted
for by these seven factors are presented in Table 1.
The factor matrix was rotated using an oblique rotation
(oblimin, with delta set to 0). Thefirsttwo factors that emerged
from this analysis clearly represented the two hypothesized
components of impulsivity. The 11 items that loaded over .30
on Factor 1 had all been written to tap functional impulsivity,
whereas the 12 items that loaded over .30 on Factor 2 had all
been written to tap dysfunctional impulsivity (See Tables 2 and
3). No item loaded over .30 on both Factor 1 and Factor 2.
The items loading on the third factor appeared to reflect a
concern about making mistakes. The items loading on the
fourth factor seemed to represent the tendency to dislike jobs
that require careful attention. The items loading on the fifth
factor appeared to concern the tendency to plan ahead, to con-
sider the long-range consequences of actions. The items loading
on the sixth factor seemed to reflect the tendency to pause be-
fore making decisions. And the items loading on the seventh
factor appeared to represent a dislike of acting quickly in gen-
eral, whether the actions involve decisions or not.
Factor 1 (Functional Impulsivity) and Factor 2 (Dysfunc-
tional Impulsivity) correlated .07. Functional Impulsivity cor-
related -.35 with Factor 6 (Pausing Before Decisions). Dys-
functional Impulsivity correlated .33 with Factor 4 (Disliking
Jobs That Require Careful Attention) and -.57 with Factor 5
(Planning Ahead).
The 11 items that loaded above .30 on Factor 1 were com-
bined to create a Functional Impulsivity scale; the internal-con-
sistency reliability of this scale (Cronbach's alpha) was .83. The
12 items that had loaded above .30 on Factor 2 were also com-
bined to create a Dysfunctional Impulsivity scale; the alpha for
this scale was .86. The correlation between these two scales
was .22.
Discussion
This study provided evidence that functional and dysfunc-
tional impulsivity can be distinguished within the self-report
domain. Both traits appear to involve the tendency to deliberate
less than most people of equal ability before taking action. For
individuals high in functional impulsivity, this tendency has
positive consequences and is a source of pride, but for individu-
als high in dysfunctional impulsivity, this tendency leads to
difficulties. The relatively low correlation between the two traits
(.07 for the factors and .22 for the derived scales) provides evi-
dence that the two traits are sufficiently independent to be
worth examining separately.
Study 2
The aim of the second study was to clarify further the nature
of functional and dysfunctional impulsivity by examining the
relation between these two traits and other self-report personal-
ity measures. These included both measures of impulsivity and
measures of traits that had previously been found to be related
to impulsivity.
It was noted earlier that previous work on the consequences
of impulsivity for cognitive functioning appears to have empha-
sized the dysfunctional aspects of impulsivity. Certain self-re-
port measures of impulsivity also seem to reflect this bias. For
example, the Narrow Impulsivity Scale (Narrow-I; S. B. G.
Eysenck & Eysenck, 1978) is correlated more highly with psy-
choticism than it is with extraversion (S. B. G. Eysenck & Eys-
enck, 1978). It is less clear for other self-report measures of im-
pulsivity whether they reflect functional or dysfunctional im-
pulsivity to a greater degree. Gerbing et al.'s (1987) analysis of
a large pool of items drawn from major impulsivity scales iden-
tified three second-order factors; two of these factors, (a) Spon-
taneous and (b) Carefree, seem to reflect functional impulsivity,
whereas the third, (c) Not Persistent, seems closer to dysfunc-
tional impulsivity. One aim of the present study, then, was to
determine directly the extent to which previous self-report
measures of impulsivity have focused on the positive versus the
negative aspects of this trait.
A second question addressed by the present study was
whether the two types of impulsivity bear different relations to
other traits known to be related to impulsivity. To the extent
that functional and dysfunctional impulsivity are distinct traits,
their respective patterns of correlations with other traits ought
to be different.
 FUNCTIONAL AND DYSFUNCTIONAL IMPULSIVITY 97
Table 2
Items Loading Primarily on Factor 1, Functional Impulsivity, in Study 1
Factor 1 Factor 2
Item loading loading
I don't like to make decisions quickly, even simple decisions, such
as choosing what to wear, or what to have for dinner. -.79 -.06
I am good at taking advantage of unexpected opportunities,
where you have to do something immediately or lose your
chance. .64 -.10
Most of the time, I can put my thoughts into words very rapidly. .38 .03
I am uncomfortable when I have to make up my mind rapidly. -.77 -.07
I like to take part in really fast-paced conversations, where you
don't have much time to think before you speak. .72 .15
I don't like to do things quickly, even when I am doing something
that is not very difficult. -.80 .02
I would enjoy working at a job that required me to make a lot of
split-second decisions. .54 .18
I like sports and games in which you have to choose your next
move very quickly. .57 .19
I have often missed out on opportunities because I couldn't make
up my mind fast enough. -.36 .10
People have admired me because I can think quickly. .60 .0!
I try to avoid activities where you have to act without much time
to think first. -.47 .07

Method
Subjects
The subjects in this study, again, were undergraduates who partici-
pated to fulfill a course requirement. There were 188 subjects (78 male
and 110 female).
Measures
A questionnaire was created containing the Functional and Dysfunc-
tional Impulsivity scales, along with 23filleritems. This questionnaire
will be referred to as the Impulsivity Inventory.
Four impulsivity scales used in previous research were administered
in this study. One of these scales was developed by Revelle et al. (1980)
on the basis of a factor analysis of the Eysenck Personality Inventory
(EPI) Extraversion scale and will be referred to here as the EPI-I Scale.
This impulsivity scale, and others derived in the same manner from the
EPI, have been the most widely used self-report measures in research
on impulsivity and information processing. The other impulsivity scales
administered were the Barratt Impulsivity Scale, Version 5 (BIS-5; Bar-
ratt, 1965), the Narrow-I (S. B. G. Eysenck & Eysenck, 1977), and the
Personality Research Form (PRF) Impulsivity scale (PRF-I; Jackson,
1967).
The scales used to measure impulsivity-related traits were ones whose
validity had been established in previous work: (a) the General Activity
scale of the Guilford-Zimmerman Temperament Survey (GZTS; Guil-
ford, Zimmerman, & Guilford, 1976), a measure of overall activity
level; (b) the Venturesomeness Scale (S. B. G. Eysenck & Eysenck,
1978), a measure of adventurousness and risk taking; (c) the Boredom
Proneness Scale (Farmer & Sundberg, 1986); (d) the Rhathymia scale
of the GZTS, a measure of enthusiasm and optimism; (e) the Play scale
of the PRF (Jackson, 1967), a measure of the degree to which the indi-
vidual adopts a light-hearted, easy-going attitude toward life; (f) the Or-
der scale of the PRF, a measure of the individual's concern about keep-
ing materials methodically organized; (g) the Cognitive Structure scale
of the PRF, a measure of the desire to make decisions based on definite
knowledge or probabilities; and (h) the Sociability scale of the EPI (Re-
velle et al., 1980), a measure of the person's preference for social versus
solitary activities.
Procedure
Subjects were administered these scales over the course of two ses-
sions, lasting about 1 hr each. All subjects received the Impulsivity In-
ventory at the first session. The other scales were administered in ran-
dom order, so that for a given scale, approximately half the subjects
completed it during thefirstsession, and half the subjects completed it
during the second session.
Results
Impulsivity Scales
The Functional and Dysfunctional Impulsivity scales showed
a correlation of .23 (p < .05) with each other. The internal-
consistency reliability of the Functional Impulsivity scale
(Cronbach's alpha) was .74 and that of the Dysfunctional Im-
pulsivity scale was .85.
The correlations between the different impulsivity scales are
presented in Table 4, along with the reliabilities of these scales.
Both the Functional and the Dysfunctional Impulsivity scales
correlated significantly with all of the other impulsivity scales
(all ps < .01). The Dysfunctional Impulsivity scale correlated
significantly more highly with EPI-I, /(185) = -3.26, p < .01;
Narrow-I, /(185) = 7.65, p < .001; and PRF-I, /(185) = 7.63,
p < .001. The two impulsivity scales showed an identical level
of correlation with the BIS-5, <(185) = - . 4 6 . The correlations
between the impulsivity scales and the scales measuring other
traits are presented in Table 5. All of the correlations in the
table are statistically significant (p < .05).
Comparisons between the correlations in Table 5 showed that
the Functional Impulsivity scale correlated significantly more
highly with the Venturesomeness, ^185) = 2.43, p < .05, and
98 SCOTT J. DICKMAN

Table 3
Items Loading Primarily on Factor 2, Dysfunctional Impulsivity, in Study 1
Factor 1 Factor 2
Item loading loading

I will often say whatever comes into my head without thinking first. .03 .72
I enjoy working out problems slowly and carefully. -.23 -.43
I frequently make appointments without thinking about whether I
will be able to keep them. .15 .53
I frequently buy things without thinking about whether or not I can
really afford them. -.04 .66
I often make up my mind without taking the time to consider the
situation from all angles. .22 .64
Often, I don't spend enough time thinking over a situation before I
act. .15 .85
I often get into trouble because I don't think before I act. .04 .79
Many times the plans I make don't work out because I haven't gone
over them carefully enough in advance. .18 .75
I rarely get involved in projects without first considering the poten-
tial problems. -.11 -.65
Before making any important decision, I carefully weigh the pros
and cons. -.20 -.46
I am good at careful reasoning. -.10 -.52
I often say and do things without considering the consequences. -.003 .82

Rhathymia, f(185) = 3.63, p < .001, scales. The Dysfunctional pulsivity factor did not correlate with either of these two factors;
Impulsivity scale correlated significantly more highly with the both of these characteristics are likely to have negative conse-
Order, ;(185) = -2.62, p < .01, and Cognitive Structure, quences for the individual.
?(185) = -2.78, p < .01, scales. Both functional and dysfunctional impulsivity showed sig-
nificant, but small, correlations with the traits of boredom
Discussion proneness and sociability. The two types of impulsivity did not
differ in the strength of their correlations with these two traits.
This study provided additional support for the usefulness of It might be noted that these patterns of correlation provide
making the distinction between functional and dysfunctional some evidence concerning the relationship between the two
impulsivity. As in the first study, the correlation between the types of impulsivity identified here and thefivebroad personal-
Functional and Dysfunctional Impulsivity scales was rela- ity traits identified by Norman and others (e.g., Norman, 1963).
tively low. Costa and McCrae (1988) have investigated the relation be-
In addition, there were different patterns of correlations be- tween the scales of the PRF and measures of these five broad
tween each type of impulsivity and the other personality traits traits. The PRF scales that were found to be most closely associ-
studied. Functional impulsivity was more closely associated ated with functional impulsivity in the present study were ones
with enthusiasm (i.e., rhathymia), adventurousness, and activ- that Costa and McCrae found to be associated with the broad
ity than was dysfunctional impulsivity. The closer association trait of extraversion. In contrast, the PRF scales found here to
between functional impulsivity and these traits could explain be most closely associated with dysfunctional impulsivity were
why functional impulsives are more likely to report benefiting ones that Costa and McCrae found to be associated with the
from their impulsivity than are dysfunctional impulsives. En- broad trait of conscientiousness. This provides additional sup-
thusiastic, active individuals who are willing to take risks are
likely to be very productive; in such individuals, the sheer quan-
tity of their output could well compensate for the high number
of errors in that output. Table 4
Alpha Reliabilities and Correlations Between the
Dysfunctional impulsivity was more strongly associated with Impulsivity Scales in Study 2
disorderliness (i.e., the low end of the Order scale) than was
functional impulsivity. And dysfunctional impulsivity was Impulsivity measures 1 2 3 4 5 6
more strongly associated with the tendency to ignore hard facts .14
1. Functional Impulsivity .74 .23 .34 .30 .23
when making decisions (i.e., the low end of the Cognitive Struc- 2. Dysfunctional Impulsivity .85 .51 .26 .73 .67
ture scale). Traits like disorderliness and a lack of concern about 3. EPI Impulsivity .23 .63 .71 .65
hard facts are likely to exacerbate the problems associated with 4. BIS-5 34 .65 .64
acting without forethought, rather than compensating for those 5. Narrow Impulsivity .65 .83
problems. This is consistent with the finding in Study 1 that 6. PRF Impulsivity .21
the Dysfunctional Impulsivity factor correlated with a factor Note. Italicized entries (in the diagonal) represent alpha reliabilities for
tapping a dislike ofjobs requiring careful work and with a factor the scales. EPI = Eysenck Personality Inventory; BIS-5 = Barratt Im-
tapping a dislike of planning ahead, whereas the Functional Im- pulsivity Scale, Version 5; PRF = Personality Research Form.
 FUNCTIONAL AND DYSFUNCTIONAL IMPULSIVITY
Table 5
Correlations Between the Present Impulsivity Scales and Measures
oflmpulsivity-Related Traits in Study 2
Measures of impulsivity-related traits
Present impulsivity
measures GA Vent BP
Functional .32 .44 .11
Dysfunctional .21 .29 .11
Note. GA = General Activity; Vent = Venturesomeness; BP = Boredom Proneness; Rha = Rhathymia;
Ord = Orderliness; CS = Cognitive Structure; Soc = Sociability.
port for the importance of distinguishing between these two
types of impulsivity.
The present data also indicated that existing global self-re-
port measures of impulsivity tap both functional and dysfunc-
tional impulsivity. Although most of these global measures ap-
pear to tap dysfunctional impulsivity somewhat more than they
do functional impulsivity, either of these two types of impulsiv-
ity could account for a particular association between a global
impulsivity measure and another personality or cognitive vari-
able.
Study 3
Having demonstrated that functional and dysfunctional im-
pulsivity can be discriminated within the self-report domain,
the next, andfinal,step in this sequence of studies was to deter-
mine which of these two types of impulsivity accounts for the
previousfindingof an overall relationship between impulsivity
and the speed and accuracy of certain perceptual processes. As
was noted earlier, the available evidence supports the view that
individual differences in impulsivity account for the overall re-
lation between extraversion and information processing. Thus,
it was expected that thefindingsof this study would represent
a first step in explaining the large body of research on extraver-
sion and information processing.
It was not clear from previous work which of the two types
of impulsivity distinguished here, functional or dysfunctional,
would account for the relationship between broad measures of
impulsivity and measures of cognitive functioning. Both types
of impulsivity correlated significantly with the measure of im-
pulsivity used in most previous work on impulsivity and infor-
mation processing, the EPI-I. And both types of impulsivity in-
volve a self-reported tendency to act with relatively little fore-
thought. Thus, on an a priori basis, it would seem that either
type of impulsivity, or both, could account for the previous
findings of a relationship between the EPI-I and cognitive func-
tioning.
Study 3 was also designed to provide data bearing on the rela-
tionship between impulsivity and ability. Perhaps functional
impulsives proceed more rapidly than most people simply be-
cause they are very bright; because of their greater intelligence,
they can proceed more rapidly than others without being any
less accurate. Study 3 was designed to provide both speed and
accuracy data, making it possible to determine whether high
functional impulsives were less accurate than other individuals,
as well as being faster.
99
Rha Play Ord CS Soc
.36 .17 .24 .28 .22
.12 .22 -.46 -.51 .15
Method
Measures and Materials
The Impulsivity Inventory used in the previous study was used here
as well. The cognitive task used in this study had previously been shown
to be sensitive to individual differences in the speed and accuracy of
basic perceptual processes (Cooper, 1976; Cooper & Podgorny, 1976;
Dickman & Meyer, 1988). Subjects examined two complex geometric
figures, presented side by side, and decided whether they were identical.
Eachfigurewas created by removing one or more As at random from
the periphery of an array of Xs that was 5 Xs high and 10 Xs wide. To
enhance the goodness of thesefigures,the array was compressed verti-
cally so that adjacent Xs touched each other both horizontally and verti-
cally.
As in previous work, pairs of both simple and complexfigureswere
used. To construct the simple figures, 8 Xs were removed at random
until thefigurecontained between 18 and 22 corners. To construct the
complex figures, 17 Xs were removed at random until thefigurecon-
tained between 38 and 42 corners. (See Dickman & Meyer, 1988 for
examples offiguresof this type.)
Thefigurepairs were printed in booklets, 112figurepairs to a book-
let. There were two booklets, one containing only pairs of simple figures,
and the other containing only pairs of complexfigures.Thefigurepairs
were in random order within each booklet. Subjects indicated whether
the twofiguresin each pair were the same or different by marking one
of two circles near the figures.
Procedure
Pretesting. In an initial screening session, 324 subjects completed the
Impulsivity Inventory, in groups of 10 to 20. Again, the subjects in this
study were undergraduates serving for course credit. This session lasted
about 20 min.
The distribution of scores for each of the two impulsivity scales was
divided into thirds. Individuals in the upper third of the distribution for
each impulsivity scale were considered to be high on that trait, whereas
individuals whose scores fell in the lower third were considered to be
low on the trait.
Subjects for the main session of Study 3 were drawn from the four
groups of subjects representing the possible combinations of high and
low scores on the two impulsivity scales. Due to scheduling problems,
not all of the eligible subjects were able to participate in the main ses-
sion; the percentage of eligible subjects who participated in this session
was very similar for the four groups. There were a total of 217 subjects
in the main session, including 75 subjects who were low in both func-
tional and dysfunctional impulsivity, 38 subjects who were low in func-
tional impulsivity and high in dysfunctional impulsivity, 48 subjects
who were high in functional impulsivity and low in dysfunctional im-
pulsivity, and 56 subjects who were high in both functional and dysfunc-
tional impulsivity.
100 SCOTT J. DICKMAN
Table 6
Means for Time per Item (TPI), Error Rate (ER), and Score in Study 3
Dysfunctional impulsivity
Low
Functional
impulsivity TPI ER
Low
Simple figures 8.3 .07
Complex figures 10.1 .13
High
Simple figures 7.7 .11
Complex figures 9.3 .15
Main session. In the main session, which lasted about 30 min, the
two figure-comparison tasks were administered to subjects in groups of
2-5 individuals. Each session began with a practice set of figures. Sub-
jects were given 2 min to compare 27 pairs of figures, printed on a single
page. The practice figures were then scored, and subjects were given
feedback on their performance.
The subjects then completed the two test booklets. They were allowed
11.5 min to work on each of the two booklets; this amount of time was
chosen so that no subject was able to complete a booklet. There was
a 5-min break between booklets. The order of the two booklets was
counterbalanced within each of the four subject groups separately. Sub-
jects were instructed not to skip any figures and not to erase mistakes.
Results
The number of items completed by each subject was trans-
formed into average time per item (TPI), in seconds. An overall
score for the task was also computed by totaling the number of
correct responses the subject made. Analyses of variance (AN-
OVAS) were carried out on TPI, error rate (ER), and score. The
regression model was used rather than the unweighted means
model because it was thought that the unequal sizes of the four
subject groups reflected the relative sizes of these groups in the
population as a whole, rather than being an artifact of the exper-
imental procedure. The w2 statistic was used to determine the
percentage of variance accounted for by the significant effects.
For all three ANOVAS, there were three factors. The two be-
tween-subjects factors were Functional Impulsivity (low, high)
and Dysfunctional Impulsivity (low, high). The within-subject
factor was Figure Complexity (low, high). The means for TPI,
ER, and score are presented in Table 6. All means presented
here are weighted means, for the reasons noted earlier.
TPI
For TPI, the main effect of figure complexity was significant,
^(1,213)= 159.21, MSE = 262.55, p <.001, <o = .43, with the
complex figures showing a longer average TPI (9.83) than the
simple figures (8.15). This provides evidence that the figure
complexity manipulation was successful. The main effect of
functional impulsivity was significant, F(l, 213) = 11.60,
MSE = 93.28, p < .001, <JI = .05; high functional impulsives
had a shorter average TPI (8.55) than low functional impulsives
(9.40). The main effect of dysfunctional impulsivity was not
significant, F{ 1,213)= 1.50, MSE = 12.05,/> = .22.
High
Score TPI ER Score
63.52 9.1 .09 57.87
48.91 10.5 .13 47.97
65.40 7.7 .10 65.96
51.75 9.5 .15 50.71
None of the interactions were significant. These included
Functional Impulsivity X Dysfunctional Impulsivity, F(l,
213) = .69, MSE = 5.52, p = .41; Functional Impulsivity X Fig-
ure Complexity, F( 1,213) = . 18, MSE = .29, p = .68; Dysfunc-
tional Impulsivity X Figure Complexity, F(l, 213) = .14,
MSE = -23, p = .71; and Functional Impulsivity X Dysfunc-
tional Impulsivity X Figure Complexity, F(l, 213) = 1.11,
MSE= 1.83,/7 = .29.
Error Rate
The findings for ER paralleled those for TPI. The main effect
of figure complexity was significant, F( 1,213) = 116.05, MSE -
.2X61,p< .001,(0 = .35; the average ER on the complex figures
(.14) was greater than that for the simple figures (.09). The main
effect of functional impulsivity was significant, F(\, 213) =
6.18, MSE = .0495, p < .01, &> = .03, with high functional im-
pulsives showing a higher average ER (.13) than low functional
impulsives (.10). The main effect of dysfunctional impulsivity
was again nonsignificant, F(l, 213) = .12, MSE = .0099,
= .n.
P
Just as for TPI, none of the interactions were significant.
These included Functional Impulsivity X Dysfunctional Impul-
sivity, F( 1,213) = . 17, MSE = .0014, p = .68; Functional Impul-
sivity X Figure Complexity, F( 1, 213) = .35, MSE = .0066, p =
.55; Dysfunctional Impulsivity X Figure Complexity, F(l,
213) = .64, MSE = .0012, p = .43; and Functional Impulsiv-
ity X Dysfunctional Impulsivity X Figure Complexity, F{ 1,
213) = 2.55, MSE = .0048,/? = .11.
Score
For score, the main effect of figure complexity was again sig-
nificant, ^ 1 , 2 1 3 ) = 272.38, MSE = 18202.65, p < .001, co =
.56; subjects on the average had higher scores for the simple
figures (63.58) than for the complex figures (49.84). The main
effect of functional impulsivity was significant, F( 1, 213) =
5.66, MSE = 1544.42, p < .05, w = .03; the average score for
high functional impulsives (65.70) was higher than for low func-
tional impulsives (61.62). The main effect of dysfunctional im-
pulsivity was not significant, F(l, 213) = 1.16, MSE = 317.39,
p = .28.
The interaction between functional and dysfunctional impul-
sivity was nonsignificant, F{1, 213) = .88, MSE = 238.84, p =
 FUNCTIONAL AND DYSFUNCTIONAL IMPULSIVITY
.35, as were the Functional Impulsivity X Figure Complexity,
F(l,213) = 1.84,Af5r£= 122.88,p= .18, and the Dysfunctional
Impulsivity X Figure Complexity interactions, F{1,213) = .93,
MSE = .61.91, p = .34. The Functional Impulsivity X Dysfunc-
tional Impulsivity X Figure Complexity interaction ap-
proached significance, F\l, 213) = 3.82, MSE = 255.15, p =
.06. High functional impulsives had higher average scores than
low functional impulsives in all conditions; this difference was
considerably (about four times) greater when subjects were also
high in dysfunctional impulsivity and the task involved the sim-
plefigures(see Table 6).
Discussion
This study provided evidence for an association between
functional impulsivity and the speed and accuracy with which
subjects carry out the processes involved in the comparison of
visual stimuli. Dysfunctional impulsivity showed no such rela-
tion to these processes.
The fast, relatively inaccurate, information-processing strat-
egy adopted by functional impulsives in this study did in fact
appear to be functional under the conditions of the experiment.
It yielded higher scores, that is, more correct answers within
the allotted time, than the slower, more accurate style of low
functional impulsives. Although subjects in the present experi-
ment were not told explicitly that the criterion of good perfor-
mance was the total number of correct responses, it seems likely
that they assumed this to be the case, since it is almost invari-
ably true for the type of time-limited paper-and-pencil tests
with which students are most familiar, that is, course examina-
tions.
It is not clear why dysfunctional impulsivity failed to show a
relationship with speed or accuracy in the present study. One
possibility is that dysfunctional impulsives adopt rapid, inaccu-
rate information-processing strategies only under certain cir-
cumstances. For example, it may be that dysfunctional impul-
sives have a particularly difficult time carrying out slow, me-
thodical information processing when they are under stress.
From this point of view, the rapid error-prone information pro-
cessing characteristic of dysfunctional impulsives reflects a
breakdown in the control of information processing. This ex-
planation resembles one offered by Block, Block, and Harring-
ton (1974) to account for the fast, inaccurate performance of
certain children on the Matching Familiar Figures Test (Kagan,
Rosman, Day, & Phillips, 1964); that test, like the global impul-
sivity measures examined in Study 2, probably taps both func-
tional and dysfunctional impulsivity. From this perspective, the
failure tofindan effect of dysfunctional impulsivity in the pres-
ent study would be due to the fact that subjects did not perceive
the present experimental situation to be particularly stressful.
The complex interaction between Functional Impulsivity,
Dysfunctional Impulsivity, and Figure Complexity that ap-
proached significance for score is difficult to interpret. It may
be that individuals who are both low in functional impulsivity
and high in dysfunctional impulsivity have particular difficul-
ties in carrying out tasks, such as comparing simple figures,
where very rapid information processing is the optimal strategy.
It is worth noting that an earlier study using the same figures
that were used in the present study found a significant interac-
tion between impulsivity and figure complexity of a different
101
sort; this interaction was significant for both speed and accu-
racy (Dickman & Meyer, 1988). In the earlier study, high impul-
sives slowed down less than other subjects when thefigureswere
complex compared to when they were simple, and they showed
a correspondingly greater increase in error rate on the complex
figures. In that study, the EPI-I was used to measure impulsivity,
and this scale, according to Study 2, taps both functional and
dysfunctional impulsivity. Thus, it is possible that the earlier
experiment was for some reason more stressful for subjects and
that this caused dysfunctional impulsivity to affect perfor-
mance more than it did in the present study. According to this
view, the stress in that experiment interfered with the ability
of dysfunctional impulsives to adopt the slow, careful strategy
necessary for comparing the complex figures.
Whatever the reasons for the failure of high and low dysfunc-
tional impulsives to differ in their performance in the present
study, this study did clearly show that functional and dysfunc-
tional impulsivity bear different relations to important aspects
of cognitive functioning. Thisfinding,in turn, provides strong
support for the usefulness of distinguishing between these two
types of impulsivity.
Finally, Study 3 also provided evidence bearing on the rela-
tion between functional impulsivity and cognitive ability. The
possibility was raised earlier that high functional impulsives
might act with less deliberation than other people simply be-
cause they are especially bright. According to this view, their
greater intelligence allows these individuals to proceed more
rapidly than most people, without being any less accurate. The
present data, however, argue against this explanation; in the
present study, the greater speed of functional impulsives was in
fact accompanied by a significant cost in accuracy.
General Discussion
The three studies described here provide considerable evi-
dence for the usefulness of distinguishing between functional
and dysfunctional impulsivity. These two types of impulsivity
differ in both their personality and their cognitive correlates. In
Study 1, functional and dysfunctional impulsivity emerged as
distinct factors in an analysis of self-report impulsivity items. In
Study 2, these two types of impulsivity proved to bear different
relation to other personality traits. And in Study 3, the two
types of impulsivity showed different relations to the way cer-
tain very basic perceptual processes are executed.
At this point, it would seem that the best way to characterize
functional impulsivity is to say that it represents the tendency
to engage in rapid, error-prone information processing (i.e., to
act with relatively little forethought) when such a strategy is ren-
dered optimal by the individual's other personality traits. Dys-
functional impulsivity appears to represent the tendency to en-
gage in rapid, error-prone information processing because of
an inability to use a slower, more methodical approach under
certain circumstances. It was suggested earlier that it might be
stressful circumstances that interfere with dysfunctional impul-
sives' ability to engage in slow, accurate information processing;
however, this is a possibility that must be confirmed by future
research.
The present findings would seem to have several important
implications for work on impulsivity and cognitive functioning.
These findings raise the possibility that functional impulsivity
102 SCOTT J. DICKMAN
is the personality trait that is most responsible for the overall
relationship between impulsivity and information processing
and, at a broader level of analysis, between extraversion and in-
formation processing. And the present data strongly suggest
that in order to discover systematic relationships between im-
pulsivity and other personality traits and between impulsivity
and cognitive processes, it will be necessary to distinguish be-
tween functional and dysfunctional impulsivity.
References
Anderson, K. J., & Revelle, W. (1983). The interactive effects of caffeine,
impulsivity, and task demands on a visual search task. Personality
and Individual Differences, 4, 127-134.
Bakan, P. (1959). Extraversion-introversion and improvement in an au-
ditory vigilance task. British Journal of Psychology, 50, 325-332.
Barratt, E. S. (1965). Factor analysis of some psychometric measures of
impulsiveness and anxiety. Psychological Reports, 16, 547-554.
Block, J., Block, J. H., & Harrington, D. (1974). Some misgivings about
the Matching Familiar Figures Test as a measure of reflection-impul-
sivity. Developmental Psychology, 10, 611-632.
Bone, R. N. (1971). Interference, extraversion, and paired-associate
learning. British Journal of Social and Clinical Psychology, 10, 284-
285.
Cole, P., & Hartley, D. (1978). The effects of reinforcement and strategy
training on impulsive responding. Child Development, 49, 381-384.
Cooper, L. A. (1976). Individual differences in visual comparison pro-
cesses. Perception andPsychophysics, 19, 433-444.
Cooper, L. A., & Podgorny, P. (1976). Mental transformations and vi-
sual comparison processes: Effects of complexity and similarity. Jour-
nal of Experimental Psychology: Human Perception and Perfor-
mance, 2, 503-514.
Costa, P. T, & McCrae, R. R. (1988). From catalog to classification:
Murray's needs and the five-factor model. Journal of Personality and
Social Psychology, 55, 258-265.
Denny, D. (1970). Modeling effects upon the conceptual style and con-
ceptual tempo of impulsive children. Developmental Psychology, 2,
22-32.
Dickman, S. (1985). Impulsivity and perception: Individual differences
in the processing of the local and global dimensions of stimuli. Jour-
nal of Personality and Social Psychology, 48, 133-149.
Dickman, S., & Meyer, D. E. (1988). Impulsivity and speed-accuracy
tradeoffs in information processing. Journal of Personality and Social
Psychology, 54, 274-290.
Eysenck, H. J., & Eysenck, M. W. (1985). Personality and individual
differences. New \brk: Plenum Press.
Eysenck, H. J., & Levey, A. (1972). Conditioning, introversion-extra-
version and the strength of the nervous system. In V. D. Nebylitsyn
and J. A. Gray (Eds.), Biological bases of individual behavior (pp.
206-220). London: Academic Press.
Eysenck, M. W. (1974). Extraversion, arousal, and retrieval from se-
mantic memory. Journal of Personality, 42, 319-331.
Eysenck, S. B. G., & Eysenck, H. J. (1963). On the dual nature of extra-
version. British Journal of Social and Clinical Psychology, 2, 46-55.
Eysenck, S. B. G., & Eysenck, H. J. (1977). The place of impulsiveness
in a dimensional system of personality description. British Journal of
Social and Clinical Psychology, 16, 57-68.
Eysenck, S. B. G., & Eysenck, H. J. (1978). Impulsiveness and venture-
someness: Their position in a dimensional system of personality de-
scription. Psychological Reports, 43, 1247-1253.
Farmer, R., & Sundberg, N. D. (1986). Boredom Proneness—The devel-
opment and correlates of a new scale. Journal of Personality Assess-
ment, 50, 4-17'.
Gerbing, D., Ahadi, S., & Patton, J. (1987). Toward a conceptualization
of impulsivity: Components across the behavioral and self-report do-
mains. Multivariate Behavioral Research, 22, 1-22.
Guilford, J. P. (1975). Factors and factors of personality. Psychological
Bulletin, 82, 802-814.
Guilford, J. S., Zimmerman, W. S., & Guilford, J. P. (1976). The Guil-
ford-Zimmerman Temperament Survey Handbook. San Diego, CA:
Edits.
Harkins, S., & Geen, R. G. (1975). Discriminability and criterion
differences between extraverts and introverts during vigilance. Jour-
nal of Research in Personality, 9, 335-340.
Heider, E. (1971). Information processing and the modification of an
"impulsive conceptual tempo." Child Development, 42, 1276-1281.
Howarth, E., & Browne, J. A. (1972). An item factor analysis of the
Eysenck Personality Inventory. British Journal of Social and Clinical
Psychology, 11, 162-174.
Jackson, D. N. (1967). Personality Research Form Manual. Goshen,
NY: Research Psychologists Press.
Kagan, J., Pearson, L., & Welch, L. (1966). Modification of an impul-
sive tempo. Journal of Educational Psychology, 57, 359-365.
Kagan, J., Rosman, B. L., Day, D., & Phillips, W. (1964). Information
processing in the child: Significance of analytic and reflective atti-
tudes. Psychological Monographs, 78,(1, Whole No. 578).
Kagan, J., Pearson, L., & Welch, L. (1966). Modification of an impul-
sive tempo. Journal of Educational Psychology, 57, 359-365.
Loo, R. (1979). Role of primary personality factors in the perception of
traffic signs and driver violations and accidents. Accident Analysis and
Prevention, 11, 125-127.
McLaughlin, R. J. (1968). Retention in paired-associate learning re-
lated to extraversion and neurotitism. Psychonomic Science, 13, 333-
334.
Meichenbaum, D., & Goodman, J. (1971). Training impulsive children
to talk to themselves: A means of developing self-control. Journal of
Abnormal Psychology, 77, 115-126.
Nelson, W, & Birkimer, J. (1978). Role of self-instruction and self-rein-
forcement in the modification of impulsivity. Journal of Consulting
and Clinical Psychology, 46, 183.
Norman, W. T. (1963). Toward an adequate taxonomy of personality
attributes: Replicated factor structure in peer nomination personality
ratings. Journal of Abnormal and Social Psychology, 66, 574-583.
Plomin, R. (1976). Extraversion: Sociability and Impulsivity? Journal
of Personality Assessment, 40, 24-30.
Revelle, W, Humphreys, M. S., Simon, L., & Gilliland, K. (1980). The
interactive effect of personality, time of day, and caffeine: A test of the
arousal model. Journal oj Experimental Psychology: General, 109, 1-
31.
Ridberg, E., Parke, R., & Hetherington, E. (1971). Modification of im-
pulsive and reflective cognitive styles through observation of film-me-
diated models. Developmental Psychology, 5, 369-377.
Wallach, M. A., & Gahm, R. C. (1960). Effects of anxiety level and
extraversion-introversion on probability learning. Psychological Re-
ports, 7, 387-398.
Zelniker, T., Jeffrey, W, Ault, R., & Parsons, J. (1972). Analysis and
modification of search strategies of impulsive and reflective children
on the Matching Familiar Figures Test. Child Development, 43,321-
325.
Received August 15,1988
Revision received March 13,1989
Accepted June 7,1989 •