Limits...
Imitation by combination: preschool age children evidence summative imitation in a novel problem-solving task.

Subiaul F, Krajkowski E, Price EE, Etz A - Front Psychol (2015)

Bottom Line: Yet, imitation and innovation are both necessary components of cumulative cultural evolution.Across experiments, more than 75% of children evidenced summative imitation, opening both compartments of the problem box and retrieving the reward hidden in each.Generally, learning different actions from two different models was as good (and in some cases, better) than learning from 1 model, but the underlying representations appear to be the same in both demonstration conditions.

View Article: PubMed Central - PubMed

Affiliation: The George Washington University, Washington DC, USA.

ABSTRACT
Children are exceptional, even 'super,' imitators but comparatively poor independent problem-solvers or innovators. Yet, imitation and innovation are both necessary components of cumulative cultural evolution. Here, we explored the relationship between imitation and innovation by assessing children's ability to generate a solution to a novel problem by imitating two different action sequences demonstrated by two different models, an example of imitation by combination, which we refer to as "summative imitation." Children (N = 181) from 3 to 5 years of age and across three experiments were tested in a baseline condition or in one of six demonstration conditions, varying in the number of models and opening techniques demonstrated. Across experiments, more than 75% of children evidenced summative imitation, opening both compartments of the problem box and retrieving the reward hidden in each. Generally, learning different actions from two different models was as good (and in some cases, better) than learning from 1 model, but the underlying representations appear to be the same in both demonstration conditions. These results show that summative imitation not only facilitates imitation learning but can also result in new solutions to problems, an essential feature of innovation and cumulative culture.

No MeSH data available.


Summary of error types by condition and experiment.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4585006&req=5

Figure 3: Summary of error types by condition and experiment.

Mentions: To disambiguate random errors from imitation-related errors, we included an analysis of errors based on learning condition (i.e., Baseline, 1 Model, 2 Model). Specifically, we analyzed whether there were differences in the types of errors children made across learning conditions. Children in the 1 and 2 models demonstration conditions did not make different types of errors (all Zs < 1.50, ps > 0.10, rs < 0.18, Mann–Whitney test). However, compared to Baseline, children in both demonstration conditions made significantly more demonstration-related errors (slide: Z = -3.05, p < 0.03, r = 0.43, lift errors: Z = 2.92, p < 0.03, r = 0.41) as well as one non-demonstration related error such as interacting with the wrong side of the box (wrong side: Z = -2.55, p = 0.03, r = 0.36). Learning conditions did not differ in terms of breaking the box while trying to find the stickers (destroy: Z = -1.40, p = 0.48, r = 0.20). All analyses have been corrected for multiple comparisons using Bonferroni Procedure. Results are summarized in Figure 3.


Imitation by combination: preschool age children evidence summative imitation in a novel problem-solving task.

Subiaul F, Krajkowski E, Price EE, Etz A - Front Psychol (2015)

Summary of error types by condition and experiment.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4585006&req=5

Figure 3: Summary of error types by condition and experiment.
Mentions: To disambiguate random errors from imitation-related errors, we included an analysis of errors based on learning condition (i.e., Baseline, 1 Model, 2 Model). Specifically, we analyzed whether there were differences in the types of errors children made across learning conditions. Children in the 1 and 2 models demonstration conditions did not make different types of errors (all Zs < 1.50, ps > 0.10, rs < 0.18, Mann–Whitney test). However, compared to Baseline, children in both demonstration conditions made significantly more demonstration-related errors (slide: Z = -3.05, p < 0.03, r = 0.43, lift errors: Z = 2.92, p < 0.03, r = 0.41) as well as one non-demonstration related error such as interacting with the wrong side of the box (wrong side: Z = -2.55, p = 0.03, r = 0.36). Learning conditions did not differ in terms of breaking the box while trying to find the stickers (destroy: Z = -1.40, p = 0.48, r = 0.20). All analyses have been corrected for multiple comparisons using Bonferroni Procedure. Results are summarized in Figure 3.

Bottom Line: Yet, imitation and innovation are both necessary components of cumulative cultural evolution.Across experiments, more than 75% of children evidenced summative imitation, opening both compartments of the problem box and retrieving the reward hidden in each.Generally, learning different actions from two different models was as good (and in some cases, better) than learning from 1 model, but the underlying representations appear to be the same in both demonstration conditions.

View Article: PubMed Central - PubMed

Affiliation: The George Washington University, Washington DC, USA.

ABSTRACT
Children are exceptional, even 'super,' imitators but comparatively poor independent problem-solvers or innovators. Yet, imitation and innovation are both necessary components of cumulative cultural evolution. Here, we explored the relationship between imitation and innovation by assessing children's ability to generate a solution to a novel problem by imitating two different action sequences demonstrated by two different models, an example of imitation by combination, which we refer to as "summative imitation." Children (N = 181) from 3 to 5 years of age and across three experiments were tested in a baseline condition or in one of six demonstration conditions, varying in the number of models and opening techniques demonstrated. Across experiments, more than 75% of children evidenced summative imitation, opening both compartments of the problem box and retrieving the reward hidden in each. Generally, learning different actions from two different models was as good (and in some cases, better) than learning from 1 model, but the underlying representations appear to be the same in both demonstration conditions. These results show that summative imitation not only facilitates imitation learning but can also result in new solutions to problems, an essential feature of innovation and cumulative culture.

No MeSH data available.