Over the responses to the contrastive questions. A positive score indicates a more positive assessment of pro-social than the anti-social agent, viceversa for negative scores. The error bars correspond to one betweensubject standard error above and below the mean. *p,.05. doi:10.1371/journal.pone.0088612.gthe counterbalancing factors, participant sex or any interactions between these factors (p.0.1). Sixteen toddlers occasionally produced short comments during the movie clips (30 comments in total), which we analyzed by taking into account the valence of the comments (neutral, positive or negative) and the action script concerned (human harming, object harming, human comforting or object comforting). All comments which were a description of the action (e.g. “he pushes the girl!”, “the girl falls”) were considered as neutral, those containing positive words (e.g. “he’s nice!”) as positive comments and those containing negative words (“bad !”, “he’s not nice”) as negative comments. If a toddler repeated the same comment in a same sequence (for example: “he’s bad!”) twice, we coded it as a single comment. A four by three contingency table was constructed (see Table 1), by tabulating the three types of comments across the four types of action scripts. There was a significant effect of comment type (X2 (2) = 6.2, p,.05), reflecting the fact that toddlers gave mostly negative comments, and very little positive comments. There was a significant effect of action scripts (X2 (3) = 17.2, p,.001), reflecting the fact that the human harming sequence generated the most comments. Finally, there was an interaction between these two factors (X2 (6) = 13.1, p,.05), reflecting the fact that by far the most frequent comments were negative comments produced during the human harming sequences. This result indicates overall a greater sensitivity to the negative act performed towards the little girl.Experiment 2 MethodFor the 10-month-old infants, we used the same design and stimuli as in Experiment 1, but the movies were recast in order to eliminate the Vadadustat supplier simulated kicking (the negative act was reduced to pushing). Infants were presented with the same set of actions as before, but the questionnaires were removed. Instead, at the end of the entire set of movie clips, the two agents were shown enteringEarly Social Evaluation of Human InteractionsTable 1. Number of positive, neutral or negative comments during the four types of action scripts.was blindfolded during the entire experiment. Among the 10month-old infants, 4 did not complete the presentation (unrest, technical failure), and 3 did not run the test session, leaving 47 infants to analyze (24 males, 23 females).Action script Human Comments Positive comments Neutral comments Negative comments Total Harming 0 5 12 17 Comforting 2 2 1 5 Object Harming 1 4 1 6 Comforting 1 0 1Results and discussionSeveral infants’ first attempts at reaching did not succeed either because the infant did not complete his or her gesture, or because they failed to remove the toy from the presentation arm. After an uncompleted reaching attempt, infants CV205-502 hydrochloride chemical information tended to reach towards the other object, and even subsequently to alternate between the two objects. Given such behavioral variability, we decided to code the first reaching attempt, towards one of the objects, whether or not the reaching was successful. Coders were instructed to use movements of one or both arms, as well as the orientation of body and head in order to.Over the responses to the contrastive questions. A positive score indicates a more positive assessment of pro-social than the anti-social agent, viceversa for negative scores. The error bars correspond to one betweensubject standard error above and below the mean. *p,.05. doi:10.1371/journal.pone.0088612.gthe counterbalancing factors, participant sex or any interactions between these factors (p.0.1). Sixteen toddlers occasionally produced short comments during the movie clips (30 comments in total), which we analyzed by taking into account the valence of the comments (neutral, positive or negative) and the action script concerned (human harming, object harming, human comforting or object comforting). All comments which were a description of the action (e.g. “he pushes the girl!”, “the girl falls”) were considered as neutral, those containing positive words (e.g. “he’s nice!”) as positive comments and those containing negative words (“bad !”, “he’s not nice”) as negative comments. If a toddler repeated the same comment in a same sequence (for example: “he’s bad!”) twice, we coded it as a single comment. A four by three contingency table was constructed (see Table 1), by tabulating the three types of comments across the four types of action scripts. There was a significant effect of comment type (X2 (2) = 6.2, p,.05), reflecting the fact that toddlers gave mostly negative comments, and very little positive comments. There was a significant effect of action scripts (X2 (3) = 17.2, p,.001), reflecting the fact that the human harming sequence generated the most comments. Finally, there was an interaction between these two factors (X2 (6) = 13.1, p,.05), reflecting the fact that by far the most frequent comments were negative comments produced during the human harming sequences. This result indicates overall a greater sensitivity to the negative act performed towards the little girl.Experiment 2 MethodFor the 10-month-old infants, we used the same design and stimuli as in Experiment 1, but the movies were recast in order to eliminate the simulated kicking (the negative act was reduced to pushing). Infants were presented with the same set of actions as before, but the questionnaires were removed. Instead, at the end of the entire set of movie clips, the two agents were shown enteringEarly Social Evaluation of Human InteractionsTable 1. Number of positive, neutral or negative comments during the four types of action scripts.was blindfolded during the entire experiment. Among the 10month-old infants, 4 did not complete the presentation (unrest, technical failure), and 3 did not run the test session, leaving 47 infants to analyze (24 males, 23 females).Action script Human Comments Positive comments Neutral comments Negative comments Total Harming 0 5 12 17 Comforting 2 2 1 5 Object Harming 1 4 1 6 Comforting 1 0 1Results and discussionSeveral infants’ first attempts at reaching did not succeed either because the infant did not complete his or her gesture, or because they failed to remove the toy from the presentation arm. After an uncompleted reaching attempt, infants tended to reach towards the other object, and even subsequently to alternate between the two objects. Given such behavioral variability, we decided to code the first reaching attempt, towards one of the objects, whether or not the reaching was successful. Coders were instructed to use movements of one or both arms, as well as the orientation of body and head in order to.
