Fig 1.
The first row shows the ten polygons used for the category angular, and the bottom row shows the ten smoothed stimuli used for the category curved.
Table 1.
Valence and Arousal for words selected from the ANEW database in IAT 1 and IAT 2.
Table 2.
Order of blocks and response mappings for participants who received congruent trials first (example from IAT 1).
Table 3.
Order of blocks and response mappings for participants who received incongruent trials first (example from IAT 1).
Table 4.
The response mapping in the 3 IAT dimensions.
Fig 2.
Illustration of the experimental procedure for the implicit association task (IAT) used in Experiment 1.
Above each stimulus, cue words were presented on the left and right sides of the screen according to the response mapping of that trial. The message “Wrong” appeared if participants pressed the wrong button.
Fig 3.
Panel A: D scores are plotted as a function of the 3 IAT dimensions (danger vs. valence vs. gender). In all three experiments there was a significant shift meaning that participants associated curved polygons to safe concepts, to positive words and to female names compared to the opposite. For angular polygons participants associated angular polygons with danger concepts, negative valence and male names. Error bars are SE of the mean. Panel B: means of RTs are shown for congruent and incongruent trials across the three IAT experiments (danger vs. valence vs. gender). Error bars are SE of the mean. At the bottom error rates are reported for congruent and incongruent trials.
Fig 4.
Experiment 2: Illustration of the experimental design.
The left panel shows the two tasks (compatible vs. incompatible) with stimuli range 40. The right panel shows the two tasks (compatible vs. incompatible) with stimuli range 80. In the compatible task the manikin was moved by the participant away from an angular shape (top) or towards a curved shape (bottom). In the incompatible task the manikin was moved by the participant towards an angular shape (top) and away from a curved shape (bottom). In this figure the manikin is shown always underneath the stimulus, but it was presented above the stimulus with equal probability.
Fig 5.
(A) stimuli with vertices range 40, (B) stimuli with vertices range 80. Top quadrant: RT1, participants’ RTs (y axis) as a function of Condition (compatible vs. incompatible; x axis) and Shape (angular vs. curved; separate bars). Middle quadrant: RT3, participants’ RTs (y axis) as a function of Condition (compatible vs. incompatible; x axis) and Shape (angular vs. curved; separate bars). Bottom quadrant: participants’ error rates (y axis) Condition (compatible vs. incompatible; x axis) and Shape (angular vs. curved; separate bars). Error bars are SE of the mean.