Supporting Online Material for by yaoyufang



                    Supporting Online Material for
        16-Month-Olds Rationally Infer Causes of Failed Actions
                          Hyowon Gweon and Laura Schulz

                E-mail: (H.G.); (L.S.)

                      Published 24 June 2011, Science 332, 1524 (2011)
                               DOI: 10.1126/science.1204493

This PDF file includes:

       Materials and Methods

Other Supporting Online Material for this manuscript includes the following:
available at

       Movie S1
Materials and Methods
     88 infants (54% girls) were recruited, 5 (one per condition) were excluded for failure
to perform either target behavior, resulting in a final N of 83: 28 in Experiment 1 and 55
in Experiment 2. Within-Object: 14; Between-Objects: 14; Within-Agent 1: 19; Within-
Agent 2: 17; Between-Agents: 19; mean: 16 months; range: 13 – 20 months; age-matched
across conditions (ANOVA, p = ns). Participants were replaced due to fussing, refusal
during the warm-up, parental interference, or experimental error: Within-Object: 4;
Between-Objects: 5; Within-Agent 1: 7; Within-Agent 2: 8; Between-Agents: 6.

     Three toys were built by attaching dowels (15 cm) and fake buttons (2 x 2 x 1 cm) to
plastic containers (10 cm diameter). The green toy had a hidden circuit and switch; when
the toy was on a hard surface and the fake button was pressed, the real switch depressed
and the toy played music (creating the illusion that the fake button activated the toy). A
highchair tray was covered with soft felt so the toy could not be activated on this tray.

      The child sat in a highchair. The parent sat on a stool approximately 60 cm away.
The parent was instructed not to interact with their child and only to smile and nod if the
child addressed them. During a warm-up, the experimenter encouraged the child to pull a
cloth (20 x 75 cm) to retrieve a toy fish. The near end of the cloth covered approximately
the top third of the highchair tray. Participants who neither pulled the cloth nor pointed
to the toy after three demonstrations were replaced. The cloth was reset after each pull.
      The experimenter introduced a basket containing the toys. In the Within-Agent 2
and Between-Agents conditions, a second experimenter entered and sat next to the first
experimenter. Toys were activated as described in the text. See also Figure. The red toy
was then placed on the cloth (approximately 90 cm away from the child); the child was
handed the green/yellow toy. The remaining toy(s) were removed and the
experimenter(s) walked out of the child’s sight.

     The first author and a second coder who was blind to hypotheses and conditions data
both coded the data for whether the child first tried to Change the Agent (handed the
green/yellow toy to the parent) or the Object (pointed to or pulled the cloth to get the red
toy). Coders agreed on 84 of the 88 clips (95.5%). The remaining 4 (4.5%) were recoded
by a third coder blind to hypotheses and condition; her data were used for all analyses.
The second coder coded the perseveration measures. A fourth coder, blind to hypotheses
and conditions, rated tapes on a scale of 1-7, 1 indicating no attempt by the parent to
communicate with the child, 7 indicating significant attempts. Ratings did not differ
across conditions: Within-Object: 3.07; Between-Objects: 2.64; Within-Agent 1: 3.50;
Within-Agent 2: 2.69; Between-Agents: 3.00 (ANOVA, F(4,76) = 0.63, p = ns.) Parents
and infants were free to shift positions throughout. To ensure that any differences in the
distance of the parent, the cloth, and the red toy with respect to the infant did not affect
the results, a fifth coder blind to hypotheses and conditions estimated each of these

distances in inches for every clip just prior to the first target response. None of the
measures differed between infants who changed the agent and changed the object. Parent
to child: Change Agent, 22.3 inches; Change Object, 22.6 inches; t(81) = .27, p = ns.
Cloth to child: Change Agent, 8.7 inches; Change Object, 8.7 inches, t(81) = .11, p = ns.
Child to red toy: Change Agent, 34.7 inches; Change Object, 34.6 inches, t(81) = .40, p =

Movie S1
Within-Object and Between-Objects conditions (Exp 1).


To top