Adults’ Perception of Identical and Non-identical Twin Infant Behaviour

Study of adults’ perception of identical and non-identical twin infant behaviour and the applicability of the contrast effect

Abstract

 The “contrast effect” is the tendency to believe that there is an outstanding temperamental difference between two non-identical twins, beyond what could be explained by genetic difference. This study examined the hypothesis that non-parent observers will perceive the difference in behavioural traits between two non-identical twin infants as greater than that of two identical twin infants. Twenty first-year university students were asked to partake in a between-subjects design, half of which were primed to rate the behaviour of identical twins, and half of which were primed to rate the behaviour of non-identical twins as they watched a 3-minute video of two infant baby girls playing with each other. Then, participants were instructed to rate each twin for two behavioural traits—affect trait and task orientation trait—and calculate the absolute difference between the scores of the two infants. The results showed that the two conditioned groups were not statistically significantly different. Therefore, it was concluded that the contrast effect did not apply to non-parents, a phenomenon that proved to be effective among parents in experiments in the past.

Introduction

 Many studies have been conducted on parents’ interaction with infants, and on a smaller scale, parents’ observations of their infant’s behaviours and tendencies. Such reports have been mostly subjective and well controlled, but there still lied subjective components because of individual parental biases when engaging with their own children (Pauli-Pott & et al., 2003). A key component of the biases is known as ‘attribution bias’—the tendency to place one attribution above another without an elaborate process of thinking (Crisp & Turner, R. N., 2007). This partiality is especially prevalent in research associated with parents’ perception of twin infant temperaments and has subsequently led to marginal errors interfering a true, unaffected judgement of temperaments.

Get Help With Your Essay

If you need assistance with writing your essay, our professional essay writing service is here to help!
Find out more about our Essay Writing Service

While preceding twin studies have also shown that intraclass correlation between identical twins was much higher than that between non-identical twins, indicating that monozygotic twins are temperamentally much more similar (Saudino, Cherny & Plomin, 2000), the ‘contrast effect’ has also impacted parental conception of infant behaviour. The contrast effect is described and studied as the phenomenon whereby parents of non-identical twins unknowingly exaggerate the differences between their children than what can be explained by genetic difference (Saudino, 2003).

This experiment provides an opportunity to challenge preceding research on the contrast effect and adults’ differing perception of monozygotic and dizygotic twins, and further discover how this phenomenon affects unrelated adults. Executing this experiment is also helpful because it presents further research on such observational biases and whether these biases are general or parent-specific.   

The hypothesis prior to the experiment is that participants who have been primed to believe that the twins are non-identical will rate a greater difference between the two infants than participants who have been primed to believe the contrary. Such a prediction is linear with the concept and basis of the contrast effect, thereby extrapolating that non-parents and parents are both affected by observational biases.

Method

Participants

10 male students and 10 female students, aged 18 to 23, have been randomly picked from first-year psychology. Precisely half of all students were informed that the twins in the video shown were identical on their sheets, and the other half were informed that the twins were non-identical. The participants were selected via course requirement.

Materials

 Two versions of response sheets were given out, one that included information that the twins were identical and one that included that they were non-identical. The response sheets consisted of two tables each that alluded to two trait measures and their three corresponding behaviour measures, which were adapted from Bayley’s Infant Behaviour Rating Scale (Bayley, 1969). The first table was designed to measure affect. The behaviour traits for affect were positive affect or mood, outgoing behaviour, and level of social interaction. The second table was task orientation, and behaviour traits included level of interest in toys, persistence and goal-directedness, and sustained attention. Participants viewed a 3-minute video of two 13-month old twin girls playing with each other. Then, they were asked to scale the twins’ behaviour from a scale of 0 to 6 (minimum to maximum evidence of behaviour). Participants added the total score and configured the absolute difference between the behaviour of each twin for both traits.

Design

 The study was conducted in the form of a between-subjects design. The independent variable was the two different conditions on the response sheet: identical or non-identical twins. The dependent variables were the absolute difference of scores between the twins for both affect and task orientation.

Procedure

 The participants in this study were randomly allocated to one of two different versions of a response sheet. One response sheet proposed that the twins were identical (10 participants), and the other that the twins were non-identical (10 participants). All participants were shown a three-minute video of two twin baby girls playing with each other three times. On the first showing, participants were advised to observe the babies’ behaviour lightly and familiarize themselves with the scale. On the next two viewings, participants were to rate the behaviour of each twin on both affect and task orientation, a measure that was derived from Bayley’s Infant Behaviour Rating Scale. Upon rating the behaviour of each twin, participants were instructed to rate the twins separately, sum up the total affect and orientation score, and effectively calculate the absolute difference.

Results

 Part A: Affect Trait Scale

 The mean absolute difference score for identical twins in the Affect Trait Scale was 3.4 (SD = 2.55), and the mean absolute difference score for non-identical twins was 2.5 (SD = 1.43) (see Figure 1 and Appendix B).

Figure 1: Bar graph showing the mean absolute difference scores and standard errors of the Affect Trait Scale for identical twins and non-identical twins.

The independent sample t-test showed no significant difference in the mean of the absolute difference scores for identical and non-identical twins ( $t_{\left(18\right)}=0.97,\mathit{ n}.s.)$

. The mean absolute difference score for identical and nonidentical twins were not statistically significantly different.

 Part B: Task Orientation Trait Scale

 The mean absolute difference score for identical twins in the Affect Trait Scale was 3.8 (SD = 2.20), and the mean absolute difference score for non-identical twins was 3.2 (SD = 2.15) (see Figure 2 and Appendix D).

Figure 2: Bar graph showing the mean absolute difference scores and standard errors of the Task Orientation Trait Scale for identical twins and non-identical twins.

The independent sample t-test revealed that the difference between the mean absolute scores for identical and non-identical twins ( $t_{\left(18\right)}=0.62,\mathit{ n}.s.)$

was not statistically significant under the criterion: task orientation.

Discussion

 The objective of the study was to inspect the difference in response to a video between two distinct groups, whereby one group was instructed to believe that the twins in that video were identical and the other was instructed to believe the contrary; the study aims to answer whether altering the genetic relationship between two humans will have an impact on one’s perception of the twins’ individual behaviours, thereby testing whether or not the contrast effect is applicable to observers who are non-parents. It was hypothesized that the participants who were instructed to believe that the twins were non-identical would rate the twins’ behaviours more contrastively than those who had been told the opposite.

 Yet, the results failed to support the hypothesis as no significant differences between the two groups of participants were identified. This conclusion was evident in the fact that the mean absolute difference score between the two groups were not drastically different statistically. It is also noteworthy to point out that the mean absolute difference scores were higher for identical twins than non-identical twins on both the affect trait scale and the task-orientation scale, further contradicting the original hypothesis. Therefore, the results of this study demonstrate that the contrast effect is likely more pertinent among parents than non-parents in the real world.

 The experiment included considerable limitations that may have led to the error bars such as the sample size of the experiment, the homogeneous identity of the sample, and a probable failure to recognize that the twins were either identical or non-identical. First, the sample size was limited to 20 people; such a small sample could not possibly embody the entire adult observer population, affecting the result from being generalized. Observing a larger sample size in the future would mitigate the effects of maximum and minimum outliers and lead to a more precise result. Furthermore, participants in this study were all Psychology students, who have studied the human mind for at least two months. Therefore, there was an inevitable potential for participants to question the value of the experiment and think more intuitively and perceptively than the average adult observer or parent. Future experiments can include students studying various fields, or even include a more varied age group. Finally, participants may have failed to read or consider the important information that the twins were either identical or non-identical, since it is a seemingly minute detail in a paragraph-long set of instructions. The failure to register this information would have rendered the results irrelevant to the objective of the experiment. To eradicate this uncertainty, experimenters can emphasize to two different workshop classes that the twins are either identical or non-identical, ensuring that participants are aware of the infants’ genetic makeup and relationship.

 Future research can also benefit from a greater knowledge of the participants themselves, on top of just gender orientation. Since diverse backgrounds shape the way humans think and perceive different behaviours, it may be valuable to inquire the participants’ environmental and cultural backgrounds, and with this awareness, scrutinize how these traits affect the scale scores. Further, since participants themselves have varying personality traits and behavioral temperaments, it may also be resourceful to know whether they identify themselves as extroverted or introverted and see how such perceptions of themselves affect their perceptions of the infant twins.

References

• Bayley, N. (1969). Bayley Scales of infant development. New York:Psychological Corp.
• Crisp, R. J. & Turner, R. N. (2010). Essential social psychology (2nd ed.). London: SAGE.
• Pauli-Pott, U., Mertesacker, B., Bade, U., Haverkock, A., & Beckmann, D. (2003). Parental perceptions and infant temperament development. Infant Behavior and Development, 26(1), 27-48. doi: 10.1016/S0163-6383(02)00167-4
• Saudino, K. J., Cherny, S. S., & Plomin, R. (2000). Parent ratings of temperament in twins: explaining the ‘too low’ DZ correlations. Twin Research and Human Genetics, 3(4), 224-233. doi: 10.1375/136905200320565193
• Saudino, K.J. (2003). The need to consider contrast effects in parent-rated temperament. Infant Behaviour & Development, 26, 118-120. doi: 10.1016/S0163-6383(02)00175-3

Appendix

Appendix A: Affect trait result

Appendix B: Affect trait scale independent samples t-test

$t=\frac{3.4–2.5}{\sqrt{\frac{\left(174–\frac{{34}^2}{10}\right)+\left(81–\frac{{25}^2}{10}\right)}{9+9}(\frac{1}{10}+\frac{1}{10})}}$

$t=\frac{0.9}{\sqrt{\frac{58.4+18.5}{18}\times \frac{1}{5}}}$

$t=\frac{0.9}{\sqrt{\frac{76.9}{18}\times \frac{1}{5}}}$

$t=\frac{0.9}{\sqrt{\frac{76.9}{90}}}$

$t=\frac{0.9}{0.9244}$

$t=0.97$

According to the one-tailed hypothesis, the degree of freedom was 18. As a result, the critical value was 1.734, when the significance level was $a=0.05$

. Since the t-score, 0.97, is neither equal to nor greater than the critical value, it was not possible to reject the null hypothesis, thereby leading to the conclusion that there was no significant difference between the groups—identical and non-identical—with regards to the affect trait scale.

Appendix C: Task Orientation trait result

Appendix D: Task Orientation trait scale independent samples t-test

$t=\frac{3.8–3.2}{\sqrt{\frac{\left(188–\frac{{38}^2}{10}\right)+\left(144–\frac{{32}^2}{10}\right)}{9+9}(\frac{1}{10}+\frac{1}{10})}}$

$t=\frac{0.6}{\sqrt{\frac{43.6+41.6}{18}\times \frac{1}{5}}}$

$t=\frac{0.6}{\sqrt{\frac{85.2}{18}\times \frac{1}{5}}}$

$t=\frac{0.6}{\sqrt{\frac{85.2}{90}}}$

$t=\frac{0.6}{0.97}$

$t=0.62$

Since the one-tailed hypothesis was used, the degrees of freedom were 18, resulting in a critical value of 1.734. Because the t-value, 0.62, is neither greater than or equal to 1.734, we cannot reject the null hypothesis. Therefore, there was no significant difference between groups identical and non-identical.