Subtle Linguistic Cues Affect Children's Motivation

Full text

Short Report
Subtle Linguistic Cues Affect
Children’s Motivation
Andrei Cimpian, Holly-Marie C. Arce, Ellen M. Markman, and Carol S. Dweck
Stanford University
Are preschoolers’ reactions to setbacks influenced by whether
their successes are rewarded with generic or nongeneric praise?
Previous research has focused on the role of category-referring
generics (e.g., ‘‘Dogs are friendly’’) in shaping children’s
knowledge about natural kinds (see Gelman, 2004). Generic
sentences can, however, refer to individuals as well as catego-
ries. For example, ‘‘John is friendly’’ is generic because it reports
a general regularity—albeit about a single person—rather than
a particular fact or episode (Carlson & Pelletier, 1995). In
contrast, the nongeneric ‘‘John was friendly at the party’’ refers
to a specific past event. Generic sentences about an individual
imply that the particular behavior commented on (e.g., John
smiling warmly) stems from a stable trait (e.g., friendliness) or
skill (see Gelman & Heyman, 1999). Are children sensitive
to this subtle connotation with respect to their own behavior? If
so, then generic praise may lead children to think in trait terms,
such that later mistakes could signal a negative trait or low
ability and therefore undermine motivation (Dweck, 1999,
2006).
Preliminary support for this claim comes from a study by
Kamins and Dweck (1999). Praising the whole person (e.g., ‘‘You
are a good boy/girl’’) after success on a task fostered helpless
responses to subsequent mistakes more than praising the
process through which success was achieved (e.g., ‘‘You found
a good way to do it’’). We suggest that children’s behavior was
in part driven by the fact that the person praise was generic,
connoting a stable trait of the child, while the process praise was
nongeneric, focusing on one specific episode. Would manipu-
lating only the genericness of the praise result in similar patterns
of coping? For example, would children’s motivation be affected
differently by ‘‘You are a good drawer’’ (generic) than by ‘‘You
did a good job drawing’’ (nongeneric)? Note that these sentences
are much more similar than the person-process pairs used by
Kamins and Dweck—so similar, in fact, that adults may not be
aware of their contrasting implications and are thus likely to use
them interchangeably in interactions with children. Demon-
strating that children praised in these two ways react differently
to a challenge would be evidence for the importance of the
generic/nongeneric distinction in shaping young children’s
conceptions of their abilities.
METHOD
Twenty-four 4-year-old children (M54 years 6 months; range 5
4 years 1 month to 5 years 1 month; 12 boys) acted out several
scenarios
1
(see Kamins & Dweck, 1999, Experiment 2). Chil-
dren chose a puppet they could use to act out their part in the
scenarios; the experimenter handled a second puppet, repre-
senting a teacher. In each scenario, the teacher puppet asked
the child puppet to draw a different object. Small pipe cleaners
were used as pretend crayons; no actual drawing or pictures
were involved. The child puppet successfully completed the
requested drawing in the first four scenarios, and the teacher
puppet praised each success generically for half the children
(‘‘You are a good drawer’’) and nongenerically for the other half
(‘‘You did a good job drawing’’). The four success scenarios were
followed by two in which the child puppet made a mistake
(omitting ears on a cat and wheels on a bus) that the teacher
puppet commented on, identically in the two conditions. These
mistake scenarios allowed us to test children’s reactions to
criticism. For debriefing, the mistake scenarios were completed
successfully.
A baseline set of self-evaluation questions was given after the
third success scenario, which involved drawing an apple. The
four questions were as follows:
Question a: ‘‘Do you like the apple that you drew, or do you not
like it?’’ (answered on a 6-point scale from 1, really not like it,
to 6, really like it)
Question b: ‘‘Did what happened in the apple story make you
feel happy or sad?’’ (answered on a 6-point scale from 1, really
sad,to6,really happy)
Address correspondence to Andrei Cimpian, Department of Psy-
chology, Stanford University, Jordan Hall, Bldg. 420, 450 Serra
Mall, Stanford, CA 94305, e-mail: acimpian@psych.stanford.edu.
1
For the full script, see http://psychology.stanford.edu/acimpian/full_script.
pdf.
PSYCHOLOGICAL SCIENCE
314 Volume 18—Number 4Copyright r2007 Association for Psychological Science

Question c: ‘‘Did everything that happened in the apple story
make you feel like you were good at drawing or not good at
drawing?’’ (forced choice)
Question d: ‘‘Did everything that happened in the apple story
make you feel like you were a good boy/girl or not a good boy/
girl?’’ (forced choice)
A second set of questions was administered after the two mistake
scenarios. In addition to versions of Questions a through d,
tailored to the content of the last mistake scenario, this second
set included the following four persistence-related questions:
Question e (would the child prefer to work again on an un-
successful or successful previous drawing?): ‘‘On another day,
when you had a chance to draw one of these again, would you
want to draw the bus [unsuccessful], want to draw the tree
[successful], or want to draw the cat [unsuccessful]?’’ (forced
choice)
Question f: ‘‘If you had a chance to do something tomorrow,
would you draw or would you do something else?’’ (forced
choice)
Questions g and h (what would the child do after the teacher’s
criticism on each of the two mistake scenarios?): ‘‘Think
about the story where you drew a cat and forgot the ears. What
would you do now?’’ and ‘‘Think about the story where you
drew a bus and forgot the wheels. What would you do now?’’
(open-ended)
For Questions g and h, answers that provided a solution to the
mistakes (e.g., ‘‘Fix it’’) were coded as mastery oriented, whereas
all others (e.g., ‘‘I would walk away’’) were coded as helpless;
intercoder reliability was 100%.
RESULTS AND DISCUSSION
To compare the two conditions, we derived a composite measure
of helpless/mastery-oriented behavior by standardizing chil-
dren’s responses within each question and averaging across
questions. At baseline, before children experienced any mis-
takes, there was no significant difference between the generic-
and nongeneric-praise conditions. This result suggests that the
two types of praise were equally rewarding. However, on the
postmistake measures, children who received generic praise
exhibited significantly more helpless behavior than children
who received nongeneric praise, t(22) 52.86, p
rep
5.953, d5
1.17. This pattern was true both for children’s self-evaluation
(Questions a–d), t(22) 51.94, p
rep
5.858, d50.79, and for
their persistence (Questions e–h), t(22) 52.28, p
rep
5.903, d5
0.93. Moreover, the pattern for all eight questions went in the
predicted direction (see Fig. 1), p5.008 by a sign test, although
only for two of them—(b) and (f)—did the difference between
conditions reach significance independently.
Generic praise implies there is a stable ability that underlies
performance; subsequent mistakes reflect on this ability and can
therefore be demoralizing. When criticized, children who had
been told they were ‘‘good drawers’’ were more likely than those
who had been told they ‘‘did a good job drawing’’ to denigrate
their skill, feel sad, avoid the unsuccessful drawings and even
drawing in general, and fail to generate strategies to repair their
mistake. When asked what he would do after the teacher’s criti-
cism, one child said, ‘‘Cry. I would do it for both ofthem. Yeah, for
the wheels and the ears.’’ Children who were told they had done
a good job had less extreme emotional reactions and better
strategies for correcting their mistakes. In sum, subtle differences
Fig. 1. Children’s responses to the eight postmistake questions in the generic- and nongeneric-praise conditions.
Error bars represent 11SEM.
Volume 18—Number 4 315
A. Cimpian et al.

in the genericness of language can influence children’s concep-
tion of their abilities and their achievement motivation.
Acknowledgments—We thank Ewart Thomas, Vikram Jaswal,
Joyce Ehrlinger, and Joe Robinson for their helpful comments,
and the teachers, children, and parents at Bing Nursery School.
REFERENCES
Carlson, G.N., & Pelletier, F.J. (1995). The generic book. Chicago:
Chicago University Press.
Dweck, C.S. (1999). Self-theories: Their role in motivation, personality,
and development. Philadelphia: Psychology Press.
Dweck, C.S. (2006). Mindset: The new psychology of success. New York:
Random House.
Gelman, S.A. (2004). Learning words for kinds: Generic noun phrases
in acquisition. In D.G. Hall & S.R. Waxman (Eds.), Weaving a
lexicon (pp. 445–484). Cambridge, MA: MIT Press.
Gelman, S.A., & Heyman, G.D. (1999). Carrot-eaters and creature-
believers: The effects of lexicalization on children’s inferences
about social categories. Psychological Science,10, 489–493.
Kamins, M.L., & Dweck, C.S. (1999). Person versus process praise and
criticism: Implications for contingent self-worth and coping.
Developmental Psychology,35, 835–847.
(RECEIVED 7/13/06; REVISION ACCEPTED 9/1/06;
FINAL MATERIALS RECEIVED 9/6/06)
316 Volume 18—Number 4
Subtle Linguistic Cues Affect Motivation