Journal of Personality and Social Psychology
1989, Vol. 57, No. 2, 358-367
Copyright 1989 by the American Psychological Association, Inc.
0022-35 M/89/SOO. 75
Cognitive Person Variables in the Delay of Gratification
of Older Children at Risk
Monica Larrea Rodriguez, Walter Mischel, and Yuichi Shoda
Columbia University
The components of self-regulation were analyzed, extending the self-imposed delay of gratification
paradigm to older children with social adjustment problems. Delay behavior was related to a network
of conceptually relevant cognitive person variables, consisting of attention deployment strategies
during delay, knowledge of delay rules, and intelligence. A positive relationship was demonstrated
between concurrent indexes of intelligence, attention deployment, and actual delay time. Moreover,
attention deployment, measured as an individual diiferences variable during the delay process, had
a direct, positive effect on delay behavior. Specifically, as the duration of delay and the frustration of
the situation increased, children who spent a higher proportion of the time distracting themselves
from the tempting elements of the delay situation were able to delay longer. The effect of attention
deployment on delay behavior was significant even when age, intelligence, and delay rule knowledge
were controlled. Likewise, delay rule knowledge significantly predicted delay time, even when age,
attention deployment, and intelligence were controlled.
The self-imposed delay of gratification paradigm assesses
how long individuals wait in order to obtain a more valued but
delayed outcome versus one that is less valued but immediately
available. In this paradigm, the role of cognition and ideation
in self-imposed delay of gratification has become increasingly
clear in the young child. Preschoolers’ waiting time substan-
tially decreases when the reward objects for which they are wait-
ing are available to their view (Mischel & Ebbesen, 1970), and
they delay more effectively after they are instructed to distract
themselves from the rewards during the delay period (e.g., Mis-
chel, Ebbesen, & Zeiss, 1972; see Mischel, 1981, 1983, 1984,
for reviews). Especially important is how children ideate about
the rewards for which they are waiting (Mischel, 1974, 1981;
Mischel & Baker, 1975). For example, when children were in-
structed to focus on the arousing (consummatory, "hot") as-
pects of the rewards for which they were waiting (e.g., their
taste), their delay time decreased; in contrast, when children
were instructed to focus on the abstract (nonconsummatory
"cool") aspects of the rewards (e.g., their shape), their delay
This research was supported in part by Grants MH39349 and 39263
from the National Institute of Health to Walter Mischel and by National
Institute of Health Research Postdoctoral Fellowship Award FD06868
to Monica Larrea Rodriguez.
We would like to thank the staff and children of Wediko Children’s
Services, whose cooperation made this work possible. We are especially
grateful to Jack Wright for his invaluable cooperation in facilitating this
research in its field setting and for his generous help in data collection
and coordination of the research team. We are also grateful to Hugh
Leightman and Harry Parad, Wediko’s directors, for their general sup-
port; Mary Powers for her vital role in the daily data collection through-
out the summer; and Jan Eisenman for her involvement in the manage-
ment of the research operation.
Correspondence concerning this article should be addressed to Mon-
ica Larrea Rodriguez or Walter Mischel, Department of Psychology,
Schermerhorn Hall, Columbia University, New York, New York 10027.
time increased. Thus, effective delay within this paradigm
seems to be crucially determined by the way children deploy
their attention (e.g., Mischel & Baker, 1975; Mischel etal. 1972;
Mischel & Moore, 1980).
Although a good deal is known about how experimental ma-
nipulations such as reward exposure and cognitive transforma-
tions affect delay behavior (e.g., Mischel, 1984) as well as other
aspects of the coping process (e.g., Holmes & Houston, 1974;
S. M. Miller, 1987; S. M. Miller & Green, 1985), the relation-
ship between spontaneous attention deployment strategies and
the ability to delay gratification, using the self-imposed delay
paradigm, has not yet been investigated. Furthermore, research
examining the role of cognitive-attentional processes on delay
of gratification has not focused on the child’s directly observ-
able attention deployment: The effects of attention on delay be-
havior have been inferred only from the experimental analyses
of the delay process itself (e.g., Mischel & Baker, 1975; Mischel
et al., 1972; Mischel & Moore, 1980). Therefore, our interest
in the present study was to examine individual differences in
spontaneous attention deployment during delay. The focus was
on the child’s directly observable attention deployment and its
relationship to goal-directed delay of gratification. We were es-
pecially interested in how, in the course of the delay period, the
spontaneous attention deployment patterns of children who de-
layed to criterion differed from those of children who did not.
From the studies on the experimental analysis of the delay
process and related work, some strategies for effectively exerting
self-control have begun to emerge (e.g., D. T. Miller, Weinstein,
& Karniol, 1978; Mischel, 1984; Toner, 1981; Toner & Smith,
1977; Yates, Yates, & Beasley, 1987). Knowledge of these strate-
gies, in turn, may serve as criteria for the child’s metacognitive
awareness of the processes that allow self-control, as well as for
executive and procedural knowledge more generally (e.g.,
Brown & DeLoache, 1978; Cantor & Kihlstrom, 1987). In a
study designed to assess the child’s developing understanding
358

COGNITIVE PERSON VARIABLES 359
of the "delay rules," Mischel and Mischel (1983) found a clear
developmental trend in the children’s knowledge of self-control
strategies relevant to delay of gratification. For example, while
most 6-year-olds knew that covering the rewards helps delay,
knowledge of the value of abstracting the rewards was not evi-
dent until about age 12.
Although an orderly and progressive understanding of the de-
lay rules has been found, there has been no empirical demon-
stration of how a child’s knowledge of delay rules relate to his
or her delay behavior. Equally unknown are the links, if any,
between spontaneous attention deployment during the delay
period and the child’s knowledge and understanding of the delay
rules. Given the significant long-term cognitive and social cor-
relates of preschool delay behavior that have been demonstrated
recently (Mischel, Shoda, & Peake, 1988) it now seems espe-
cially important to investigate concurrent relations between in-
dividual differences in children’s attention deployment strate-
gies during goal-directed delay of gratification, their knowledge
and understanding of self-regulation, and their ability to delay
gratification.
Finally, in spite of the fact that deficient delay of gratification
is considered a key feature used to characterize older children
and adults with self-regulatory problems (e.g., aggressiveness,
conduct disorders, hyperactrvity), and that many self-instruc-
tional, cognitive-behavioral procedures have been devised to
treat problems of impulse control (Kanfer & Gaelick, 1986;
Karoly, 1977; Kendall, 1977; Meichenbaum & Goodman,
1971), research on delay of gratification so far has concentrated
on preschool children without known self-control problems.
The delay process in older subjects with self-regulatory and be-
havior problems remains surprisingly unstudied, although
these are the very individuals for whom attention deployment
and the ability to delay gratification are assumed to be charac-
teristically difficult. (Douglas, 1972, 1980; Ross & Ross, 1982).
Therefore, the present study extended the self-imposed delay
paradigm to a population of older subjects displaying a wide
variety of social adjustment problems, such as acting out, ag-
gression, and depression-withdrawal (see Horowitz, Wright,
Lowenstein, & Parad, 1981). These children were considered at
risk for maladaptive developmental outcomes such as adoles-
cent conduct disorders, school failure, and various psychologi-
cal dysfunctions. Most of these children could be characterized,
in varying degrees, as suffering from difficulties in impulse con-
trol and self-regulation.
In this population of at-risk children we attempted to clarify
the relationship between their knowledge of the delay rules,
spontaneous use of attentional strategies during the delay pro-
cess, and delay of gratification. We also included a measure of
verbal-intellectual ability (Peabody Picture Vocabulary Test),
because intelligence, broadly defined, is likely to have a role in
delay of gratification, self-regulatory knowledge, and attention
deployment (see Kopp, 1982; Vaughn, Kopp, & Krakow, 1984),
which should be specified. It seemed especially necessary to
measure intelligence given that no other study has examined
directly the relation between measures of intelligence and the
ability to execute self-imposed delay in the present paradigm.
Because exposure to the rewards has proved to make delay
behavior more difficult in preschoolers, its effects on the present
sample of older children also were assessed. We expected that
reward exposure would become less relevant with increasing
age: The child’s increasing knowledge of the delay rules should
clearly help overcome the impact of reward exposure (Mischel
& Mischel, 1983). Likewise, as the child develops, internal cog-
nitive-attentional processes should mediate self-regulation in-
creasingly and tend to outweigh the impact of the external situa-
tion (e.g., Mischel, 1981). However, because this is also a popu-
lation of at-risk children with self-regulatory difficulties, the
possible effects of reward exposure are less clear and require
empirical exploration.
Guided by previous findings and theorizing with younger
children, we developed the following specific expectations:
Effective attention deployment should have a direct effect on
children’s ability to delay gratification, even when controlling
for age, delay rule knowledge, and verbal-intellectual ability.
Theoretically, the importance of attention deployment in the
regulation of delay behavior has emerged from a network of
construct validity research (e.g., Mischel et al., 1972; Mischel,
1984; Vaughn, Kopp, Krakow, Johnson, & Schwartz, 1986).
The experimental findings from delay of gratification studies
with preschoolers consistently suggest that when children are
given attentional strategies to reduce the frustration of delay,
they are able to wait longer. The present study tested the hypoth-
esis that those children who use such strategies spontaneously
(i.e., when they are not externally supplied) also are likely to
delay longer. Specifically, we expected those children who wait
until they obtain the preferred, delayed outcome will display the
efficacious attention deployment pattern (distracting from the
frustratively arousing and tempting elements of the waiting sit-
uation) to a greater degree than those who terminate before re-
ceiving the preferred but delayed reward.
It is already known that children’s willingness to choose a
delayed, preferred outcome over an immediate, less preferred
one increases with age and is related to measures of intelligence
(e.g., Mischel & Metzner, 1962). We likewise expected that the
ability to sustain delay behavior itself would be positively re-
lated to age and verbal-intellectual ability. Moreover, the ability
to delay gratification should be facilitated by children’s knowl-
edge and understanding of the delay rules. Although knowledge
of the delay rules was expected to increase both with age (Mis-
chel & Mischel, 1983), and with higher verbal-intellectual abil-
ity, we also expected rule knowledge to facilitate delay of grati-
fication even when age and intelligence were controlled.
Finally, we expected that the effects of age, delay rule knowl-
edge, and verbal-intellectual ability on delay behavior would be
mediated by attention deployment. With increasing age, as they
mature cognitively and become more knowledgeable, children
should become able to deploy their attention more effectively
during the delay process, and in turn, wait longer. Similarly, to
the degree that children with higher verbal-intellectual ability
have effective attentional strategies more available, they should
be able to delay longer.
Method
Overview
To measure delay of gratification for a preferred but delayed outcome,
6- to 12-year-old children characterized as having impulsivity-related