What is the difference between imitation and mimicry

Imitation is much more sophisticated. With imitating, the voice over artist has to copy everything that has to do with what he or she is repeating. That includes the tone, phrasing, sound, motivation and delivery. An example of a voice over doing imitation are those that work as voice matchers.

You would be amazed if you could hear voice over artists who imitate. It sounds just like who or what they are imitating. If you missed the show this Saturday, listen in on Tuesday tomorrow to catch the episode again or go back to the And Sometimes Y website at the CBC to listen to the show at a later date and hear what some very intelligent people have to say about the difference between mimicry and imitation.

Imitation is a more sophisticated and deliberate means of recreating an action or a sound, consciously attempting to copy the tone, motivation, inflection, delivery, and phrasing of a particular sound or action. A good example of imitation and imitators are comedians and those who are called sound alikes and voice matchers in our industry. A good imitator can replicate an extremely close facsimile or produce an indistinguishable likeness to the voice of someone else. In the end, mimicry is similar to an educated guess whereas imitation resonates more with research and critical thinking.

The voice over industry is filled with actors who can perform a multitude of accents. Learn how to do accents with dialect coach Sammi Grant. The best TV commercial length is 30 seconds. Learn why some lengths of TV commercials are more effective, and watch some examples along the way. Second, we computed the Spearman-Brown coefficient in order to test the reliability of the automatic imitation task.

To test the relationship between mimicry and automatic imitation, we ran correlational analyses and will report both the corresponding p -values and BFs. In order to test the relationship between mimicry and automatic imitation, we correlated the mimicry score with all the scores of the automatic imitation task i. We then computed the BFs of all correlations to investigate whether the null hypothesis i. These results indicate moderate to strong evidence for the null hypothesis, which confirms that mimicry and automatic imitation are not positively correlated cf.

In further correlation analyses, we tested whether mimicry and automatic imitation correlated with the assessed personality traits and demographic data. That is, we ran three separate correlational analyses.

In the first analysis, we ran a correlation analysis between the mimicry score and all assessed personality scales as well as demographic data. In the second analysis, we correlated all reaction-time-based scores from the automatic imitation task i. Finally, in the third analysis, we calculated the same correlations for the error-rate-based scores of the automatic imitation task.

For each of the three correlational analyses, we corrected for multiple comparisons by the means of FDR [ 92 ]. An overview of all the correlations is provided in Table 3. Firstly, we correlated the reaction-time-based scores of the automatic imitation task with all assessed scales.

Secondly, we tested for gender differences in the reaction-time-based imitation measures. Thirdly, we ran the same analyses for the error-based automatic imitation scores.

While social psychologists assess mimicry, cognitive psychologists assess automatic imitation. Both forms of imitation share similarities, but also differ in important aspects. Although it is largely assumed that mimicry and automatic imitation are grounded in similar underlying processes, the assessment of mimicry and automatic imitation differs in many methodological aspects.

This raises the fundamental question to which degree these two forms of imitation are actually correlated to each other. In order to shed light on this open question, we measured mimicry as well as automatic imitation and correlated these two tasks. Moreover, we tested the relation of personality scales that have previously been reported to correlate with imitation i.

In line with past research, we detected significant imitation as well as mimicry effects. However, we did not find a significant correlation between the two tasks. This was further supported by Bayesian analyses showing that mimicry and automatic imitation were indeed not correlated. In addition, none of the assessed personality scales correlated with mimicry or automatic imitation in a meaningful way. When controlling for multiple comparisons, there was only one single significant relation, namely a correlation between the reaction time based congruency effect and the Personality Distress subscale of the Interpersonal Reactivity Index IRI; [ 52 ].

The absence of a correlation between mimicry and automatic imitation might be due to different reasons. First, it could be caused by differences in the methodological setup of the two paradigms.

While automatic imitation is measured trial-by-trial, is based on reaction times or error rates, elicits awareness in participants, and needs explicit cognitive control, mimicry is based on behavioral observations, is rather subjectively detected, remains unaware for participants, and does not need explicit cognitive control.

Although mimicry and automatic imitation did not correlate with each other, we do not argue that these two constructs are not related to each other at all. Rather, we suggest that both paradigms are part of the same construct i. That is, while mimicry measures imitation in a social context, automatic imitation measures imitation in a laboratory context instead for a similar distinction see [ 28 ]. Second, another reason for the missing correlation could be found in the reliability of the two tasks.

While the automatic imitation task produced good reliability, the mimicry task was not reliable at all. It is important to note that in line with past research on the mimicry task e. However, as far as we know, we measured for the first time the actual reliability of the mimicry task—that is the internal consistency of mimicry performance between odd and even minutes—and detected very low reliability.

Psychometric theory has shown that unreliable tasks are less likely to correlate with other tasks cf. Thus, it might well be that mimicry and automatic imitation are strongly linked to each other, but that the low reliability of the mimicry task reduces the likelihood to detect such a relation.

Despite the low reliability, it is important to note, however, that a reduced reliability does not necessarily indicate that the task is not replicable e.

Indeed, mimicry tasks have been replicated numerous times by different researchers in different labs for an overview see [ 30 ]. Since past research has not reported internal consistency of mimicry performance, we do not know how reliable mimicry tasks are in general.

But, if mimicry tasks are indeed rather unreliable, one should interpret past correlations of this task with caution. Future research may, thus, consider other tasks that assess mimicry with multiple trials in order to achieve higher internal consistency. For example, one could use tasks that measure mimicry of facial expression. It has been found that such tasks have good sensitivity to ASD related behaviors [ 96 ], and are sensitive to social variables [ 97 ].

The second finding that most of the assessed personality scales did not correlate with mimicry and automatic imitation in a meaningful way may be surprising. There might be different reasons for such a finding. First, given that unreliable tasks are less likely to correlate with other constructs cf.

Moreover, since some of the assessed scales were not very reliable either, it is also not surprising that we did not find more significant results. Second, it might be that controlling for multiple comparisons reduced statistical power and as such masked some potential effects. Indeed, when not controlling for multiple comparisons, some correlations were significant. It is important to note, however, that all correlation coefficients were very low i.

Since irrelevant effects can become significant with large samples e. Third, it might be that the assessed scales actually do not correlate with imitation. This interpretation is in line with other recent studies that did not find significant relations between imitation and different personality scales either. For instance, Butler and Ramsey [ 66 ] did not find a relation between automatic imitation and extraversion, agreeableness, disorders of social cognition i. A crucial difference between studies that found a relation with personality scales and studies that did not is the sample size.

While previous studies used rather low numbers of participants, we, but also Butler and Ramsey [ 66 ], aimed for high power. As underpowered studies in combination with publication bias increase the likelihood of false positives e.

Moreover, the finding that the automatic imitation task does not correlate with mimicry and most social related personality traits raises the question to which degree automatic imitation actually taps into social processes. While previous research suggests that automatic imitation relates to social factors such as eye contact [ ], social attitudes [ ], human-like actions e. For instance, some studies did not find increased automatic imitation in human as compared to non-human agents e.

Moreover, Butler and Ramesey [ 66 ] did not find a relationship between social components of personality traits and automatic imitation. Likewise, some researchers did not find reduced automatic imitation in individuals with autism [ 62 — 64 ]. Finally, Farmer et al. Thus, in light of these contradictory findings, more research is needed in order to draw specific conclusions about the degree to which social factors facilitate automatic imitation. Despite the lack of correlations, there are some other findings that merit further discussion.

First, we found a significant correlation between the reaction-time-based congruency effect and the Personality Distress subscale of the Interpersonal Reactivity Index IRI; [ 52 ].

While this finding goes in line with other research suggesting a positive relationship between automatic imitation and empathy [ 47 — 49 ], our effect should be interpreted with caution because this relation was only significant between one subscale of the IRI and one automatic imitation score.

Second, when not controlling for multiple comparisons, some correlations between automatic imitation and the assessed personality scales would actually have correlated significantly. Although due to our large sample size [ 98 , 99 ] the correlations that were significant at the uncorrected level should be regarded with caution, future research may follow up these correlations in order to test to which degree they are nonetheless reliable.

Third, we found a significant gender difference in automatic imitation with female participants showing stronger automatic imitation effects than male participants. While this finding is in line with previous studies e.

It might well be that differences in personality traits account for this difference. However, personality traits including empathy, autism-spectrum quotient, and traits related to self- versus other-focus cannot explain the sex-difference, as we did not find a relation between these traits and the automatic imitation task. Thus, future research may test other personality traits that are more prone to sex-differences.

First, it is possible that our null results could be explained by the quality of the assessed scales. Indeed, some subscales had rather low reliability. However, most scales that have been previously found to correlate with imitation i. Moreover, all measures that were used had been validated and are well established in the literature.

Nevertheless, future research should use additional scales and scales with more reliable long-formats. Second, one could argue that automatic imitation in the current study was confounded with spatial compatibility. To control for left-right spatial compatibility, we used a well-known method in which the stimulus hand is presented orthogonal to the response hand e. Nevertheless, a potential concern with this method is that it fails to control for orthogonal spatial compatibility [ ].

As a result, it is possible that orthogonal spatial processes introduced noise to the data that in turn masked potential correlations. Given that automatic imitation can still be observed even when spatial processes cannot contribute [ 28 ], future research should use tasks that control not only left-right, but also orthogonal spatial compatibility [ ] in order to reduce noise in the data and as such increase statistical power.

Third, as it is common in psychological research, most of our participants were female. Although 61 male subjects is atypically high for psychological research, it may still be that the lower percentage of men, as compared to women, reduced the variability within the tasks causing null effects.

In a related vein, it is important to note that all of our subjects were young psychology students. This, again, may have caused low levels of variance within the tasks, increasing the likelihood of finding null effects.

Future research should, thus, aim at assessing more intermixed samples. The present study provides novel insights into the understanding of imitation. First, we found no relationship between mimicry and automatic imitation suggesting that, despite similar underlying mechanisms, the two forms of imitation are less related to each other than initially suggested.

Second, we did not find meaningful relationships between any form of imitation and most personality traits including empathy, autism-like personality traits, and traits related to self- versus other-focus. Therefore, we suggest in line with other research i. As previous studies on the relation between personality traits and imitation tested their predictions on rather low numbers of participants, our research stresses the importance to study larger samples in order to replicate and extend previously established findings.

The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. National Center for Biotechnology Information , U. PLoS One. Published online Sep 6.

Marco Iacoboni, Editor. Author information Article notes Copyright and License information Disclaimer. Competing Interests: The authors have declared that no competing interests exist. Received Apr 4; Accepted Aug This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

This article has been cited by other articles in PMC. Abstract It is widely known that individuals have a tendency to imitate each other. Introduction Individuals imitate a wide range of different behaviors including facial expressions [ 1 ], characteristics of language [ 2 — 5 ], emotions [ 1 , 6 ], postures [ 7 ], gestures [ 8 ], complex action patterns [ 9 ] or simple movements [ 10 — 14 ]—to name just a few examples. Methodological differences between mimicry and automatic imitation Paradigms on mimicry and automatic imitation differ in many methodological aspects for an overview see Table 1.

Table 1 Differences between mimicry and automatic imitation tasks. Task characteristics Mimicry Automatic imitation Ecological validity high low Dependent variable Subjective ratings of executed actions Reaction times and error rates Awareness low high Cognitive control low high.

Open in a separate window. Theoretical similarities between mimicry and automatic imitation Although the paradigms used in research on mimicry and automatic imitation differ in several aspects, there are also important similarities with regard to the theoretical assumptions and moderators of mimicry and automatic imitation.

Theoretical assumptions Irrespective of how imitation is assessed, it is generally agreed that automatic imitation and mimicry are both based on shared representations of observed and executed actions. Moderating influences Past research on moderating influences suggests that similar factors facilitate mimicry and automatic imitation.

Present research Taken together, the literature on imitation does not offer a clear answer to the question whether mimicry and automatic imitation are correlated and to which degree personality traits moderate imitation. Method Ethics statement The study was conducted in accordance with the ethical standards of the Declaration of Helsinki and approved by the rules of the Institutional Review Board from the Faculty of Psychology and Educational Science of Ghent University.

Data availability statement The data file of the study is available from the Open Science Framework database. Participants In return for partial course credit, two hundred students from Ghent University Belgium participated in this study. Procedure After being welcomed, participants were seated in separate cubicles. Fig 1. Overview of incongruent, congruent and neutral trials in the automatic imitation task.

Personality scales In order to test the degree to which personality traits that have previously been reported to moderate imitation correlate with mimicry and automatic imitation, we administered a couple of questionnaires.

Demographic data and further personality scales Besides gender and age, we assessed further scales that had not yet been tested to moderate mimicry and automatic imitation. Mimicry Mimicry First, we ran a 2 observed video: nose touching video vs. Fig 2. Amount of performed actions in the mimicry task. Error bars represent standard errors of the mean. Reliability In a second analysis we tested the split-half reliability of the mimicry effect. Automatic imitation Latencies Fig 3 illustrates the results for the latencies.

Fig 3. Mean reaction times of the automatic imitation task separated by condition. Error rates As can be seen in Fig 4 , the results for the error rates were in line with the results for the latencies. Fig 4. Mean error rates within the automatic imitation task separated by condition. Reliability In a further analysis, we tested the split-half reliability of the automatic imitation task.

Relationship between mimicry and automatic imitation To test the relationship between mimicry and automatic imitation, we ran correlational analyses and will report both the corresponding p -values and BFs. Table 2 Intercorrelations between the mimicry score and all different automatic imitation scores. Mimicry — Congruency RT — 0.

Facilitation RT — 0. Interference RT — 0. Congruency ER — 0. Facilitation ER —