2 of 19

Key paper

Jaswal, V. K., & Neely, L. A. (2006). Adults don’t always know best: preschoolers use past reliability over age when learning new words. Psychological Science, 17(9), 757–758. (JA)

What and when do children learn from their peers?

Vikram Jaswal (UVA) & Leslie Neely (UTSA)

3 of 19

Participants

58 children—any age of 3-year-old or 4-year-old

Between subjects design: Randomly divided into 4 groups of N = 14–15:

Both reliable
Adult reliable
Child reliable
Both unreliable

Our study:

59 children—any age of 3-year-old or 4-year-old�(5 excluded for failure to train/answer to test or given pre-set inclusion criteria)

Between subjects design: Randomly divided into 4 groups:

Both reliable (N = 15)
Adult reliable (N = 15)
Child reliable (N = 17)
Both unreliable (N = 12)

4 of 19

Trial structure (orig)

sneaker!

shoe!

Shoe. Sneaker.�What is this called?

Sneaker.

Could it be called something else?

Shoe.

Did anyone say anything wrong?

No.

Training trials only:

“Corrective feedback as necessary”

5 of 19

Trial structure (ours)

sneaker!

shoe!

What did E & M say? They said: “Shoe” and “Sneaker”.�What is this called?

???

Training trials only:

“Corrective feedback as necessary”

Sneaker.

(In case of non-choice; mas 2x):�Just choose one. Is it: “Shoe” or “Sneaker”?

6 of 19

Testing phase

Training phase

shoe

airplane

cup

telephone

sneaker

plane

glass

phone

wug

blicket

dax

zimmer

zup

chroma

fips

haiger

Both reliable condition (orig stims)

7 of 19

Testing phase

Training phase

coat

jeans

cup

sofa

jacket

pants

glass

couch

peyf

darill

hayg

kraums

wesmer

yem

swuk

trobes

Both reliable condition (our stims)

8 of 19

Analysis & Results (orig)

T-test of difference from chance in each condition

ANOVA of # of “adult” selections for differences across conditions +

Tukey’s test for pairwise differences in each condition

Age & sex dropped from analysis (FN1)

9 of 19

Results (orig)

Results (ours)

10 of 19

Choices greater than chance

Both reliable:

t(13) = 2.22, p_rep = .92 (p = 0.02), d = 0.59

Reliable adult:

t(13) = 5.09, p_rep = .99 (p < .001), d = 1.36

Reliable child:

t(14) = 4.68, p_rep = .99 (p < .001), d = 1.21

Both unreliable:

t(14) < 1

Choices not greater than chance

Both reliable:

t(14) = 0.69, p = .49, d = 0.18

Reliable adult:

t(14) = 1.85, p = .086, d = 0.48

Reliable child:

t(16) = -0.637, p = .53, d = 0.15

Both unreliable:

t(11) = -0.56, p = .59, d = 0.16

11 of 19

Significant effect of condition

F(3,54) = 13.31,�p_rep = .99, (p < .001)�η² = .43

Tukey’s HSD, p < .05:

Both reliable, reliable adult > reliable child
Reliable adult > Both unreliable

No significant effect of condition

F(3,55) = 1.30,�p = 0.28,�η² = .07

Tukey’s HSD, ps ≥ .32

12 of 19