Abstract
The research topic relates to the simulation of others’ actions, or imitation. It is seen that although the act of imitation could be reasonably performed, the method by which imitations are carried out, and how imitations are really effected, cannot be fully explained.There are several theories floating and the prominent among them are the active intermodel-mapping model (AIM) and the associative sequence learning models (ASL) which have been picked up for detailed discussion in this study. By making a comparative analysis of these two methods, AIM and ASL during the course of study, as applied to opaque and transparent studies of ear and foot, it is seen that the results of both the methods differ widely.
The current studies aim to discern the forecasts made by the two reigning theories – The active intermodel-mapping model (AIM) and the associative sequence learning models (ASL). This was done through investigations of the imitations of novel opaque (ear movement) and transparent action (toe movements) Participants were required to perform these novel actions, either with, or without visual feedbacks.
Thus, the impact of visual feedback on imitation of new actions were examined.
The hypotheses to be tested are the following:
- Under AIM: No difference between feedback and non-feedback condition for either toe or ear movements. Participants should learn in both.
- Under ASL: More learning in feedback condition than non- feedback for ear movements (where there should be little or no learning).No difference between feedback conditioning for toe movements. Participants should learn in both.
The results from this study circumspectly supports ASL Model, but it is believed that are yet plenty of aspects to be discussed before a final decision is awarded.
Theories of imitations
Introduction
The aspect of imitation is a common subject in psychology, by which the observer tries to copy the body movements of another based upon observed characteristics of body movements.
“An observer can be said to imitate an instructor when:
- The observer produces behavior similar to the instructor.
- The observer’s action is caused by perception of the instructor.
- Generating the response depends on equivalence between the observer’s self-generated actions and the actions of the instructor. (Hoffman, et al. 2006, P. 299).
An important aspect regarding imitation is that when we observe the outward manifestation of movements of subjects, we observe only the outward movements and not the internal muscular activations. Thus, a pertinent question that may arise would be that in the absence of knowledge about which internal muscle groups are responsible for which types of body movements, how could we imitate body movements without knowing about the activation of internal muscle groups? This has come to be known as the correspondence problem and is a matter of great interest for behavioral psychologists and health care practitioners.
“A crucial problem in imitation is the correspondence problem, mapping action sequences of the demonstrator and the imitator agent. This problem becomes particularly obvious when the two agents do not share the same embodiment and affordances.” (Alissandrakis, 2008).
Research in this area sought to determine as to how observation of action facilitates production of movement and offers alternative solution to correspondence problem mentioned above.
There are basically two kinds of actions, transparent action and non-transparent actions. In the case of opaque actions, where there are conspicuous differences between one task performance, based on the observance of task performance of the visual feedbacks rendered by another, this imitation is rendered progressively difficult when visual signals are poor or distorted.
This essay underlines the process by which copying another person’s movement is possible. This is significant because, while it is possible to imitate another person, the method by which this is done, may not be easily explained. However, it could be said, “Imitation is based on the automatic activation of motor representations by movement observation.” (Brass and & Heyes, 2003).
This study explores the challenges for theories for the imitation and offers explanation as to how the observation of action facilitates production of machine movements under the following processes:
- The active intermodel mapping model (AIM).
- The associative sequence learning models (ASL).
The hypotheses to be tested are the following:
- Under AIM: No difference between feedback and non-feedback condition for either toe or ear movements. Participants should learn in both.
- Under ASL: More learning in feedback condition than non- feedback for ear movements (where there should be little or no learning).No difference between feedback conditioning for toe movements. Participants should learn in both.
AIM Model
The AIM Model, which is being discussed, could be seen in terms of the fact that the infant world come to identify the parts of its body with that of others. This, according to Meltzoff and Moore are the first step to initiative process. P 3. (Bird & Heyes, 2004, P.4).
The concept is that of an action used as a series of relationships between various organs of the body. The concept proceeds to cover the learning of the relationships between muscle movements and the organ relations.
It is seen that the two theories, Active intermodal mapping modal (AIM) and Associated Sequence Learning Model (ASL) have been the mainstay upon which the aspects of imitation have hinged. These theories have made significant headway into the understanding of the conceptual knowledge and empirical studies that have gone into the framework of conceptual and practical knowledge of the imitation theory in its absolute terms. Although there have been other theories in vogue relating to imitation theory, by far, it has been AIM and ASL that have held sway in research matters. It is necessary to take a fresh look at these two aspects, and how these theories have helped in the cause of understanding them in its perspective.
What is important is the manifestation of imitation in the behavior pattern of people, not in the sense of learning but in the sense of imitation performance. The practical application of imitation in the empirical sense of the term, it is necessary to focus on imitation at the level of individual behavior, responses to external and internal stimuli, and how it could be controlled and improved.
The establishment of moderating in attitudes and behavior patterns involving imitations could be established and how the greater interests of society vis-à-vis the individual could be made. It needs to be seen how the psychological and neuropsychic aspects of imitating could be analyzed, researched in its fullest depth by competent professionals, and necessary inferences drawn.
Research on imitation: It is seen that imitation learning occurs mostly by chance. These are however, reinforced by observers when activities spread, but it is seem that on most occasions the imitating is spontaneous and the child does not wish to imitate the parent intentionally. Thus, for understanding cognitive approaches, it is not only necessary to refer to one’s body movements, but see it in relationships with other bodies and its activities. (Meltzoff, & Prinz, 2002). Then only, the correct understanding of the concept of imitating can be gained and understood in its proper perspective.
AIM: During the year 1997, two research scientists, Meltzoff and Moore had created a model that explains how an infant performs and achieves factors like:
- Recognition of facial features of demonstrator.
- Establish link between perceived features and own features.
- Form a desired expression from this communication.
- Establish a configuration.
- Usage of configuration to judge own’s success.
The AIM model is important as it is used as a guide for the implementation of the social humanoid robots. (Mjølne, 2005).
Method
Participants
Twenty paid volunteers (X men, mean age X+/-X) who were both unable to move their right ear and unable to move their big toe sideways was selected for participation in this study. Participants were randomly allocated to feedback and no feedback groups. Half of the participants in each group performed ear movements first; the other half of participants performed toe movements first. Ear and toe movement blocks were completed within 24 hours of each other, either on the same day (X participants) or on two subsequent days (X participants).
Materials and apparatus
Subjects sat with their chin on a chinrest at a table at a viewing distance of approximately X cm in front of a laptop computer screen (stats). All stimuli presented on this screen were shown on a black background.
To demonstrate the ear and toe movements, videos of a model making these movements were shown for the duration of the demonstration interval. Each demonstration video consisted of 50 frames. The first 12 frames showed the model at rest. Frames 13 – 37 showed the movement; the highest movement amplitude (largest displacement of ear or toe) was reached at frame 23 and lasted until frame 27. The last 13 frames showed the model at rest after the movement. Demonstration videos were shown at a rate of 25 frames per second, rendering each video a duration of 2000 ms. The size of the model’s ear a with respect to the model’s toe was adjusted so that the largest displacement of the outer edge of the ear matched that of the outer edge of the big toe (7 mm, work out visual angle) (see Fig.1b). The model’s foot at rest (see Fig.1a top panel) occupied X(10.25cm) of the horizontal visual angle and X (17cm) of the vertical visual angle. The model’s ear at rest (see Fig.1a bottom panel) occupied X (7cm) of the horizontal and X (12cm) of the vertical visual angle.
Subjects in the no feedback group were shown the first frame of each ear and toe movement video (depicting the ear or toe at rest, see Fig.1a) for the duration of the movement interval. The frame shown at 50% of the size each video was shown during the demonstration interval. Subjects in the feedback group were shown a live video stream of their ear or toe for the duration of the movement interval. This was done using custom-made streaming software, which relayed the currently captured video to the screen. The size of the window in which the live video stream was shown was approximately 9 cm across and 7 cm high (work out visual angle) and the frame was chosen to show a similar section of the ear or foot of the subject as the demonstration video showed of the ear or foot of the model.
Design and Procedure
Each subject completed four blocks of ear movements, and four blocks of toe movements in a separate session. The first block consisted of 20 pre-training test trials; the second block consisted of 40 training trials. The third block consisted of 20 post-training test trials, and the fourth block consisted of 20 verbal instruction trials.
Each trial in pre- and post-training tests and in training began with the presentation of the demonstration video showing the model performing an ear or toe movement, repeated five times. As each movement video lasted 2000 ms, the complete demonstration interval lasted 10 seconds. In verbal instruction blocks, each trial began with the presentation of the verbal instruction “When you see the GO signal move you ear (foot) five times” shown in white ink for 10 seconds. After a pause of 1400 ms, the word “GO” was presented in white ink for 200 ms, indicating the start of the movement interval, which lasted 10 seconds.
For one-half of trials in pre- and post-training blocks as well as in verbal instruction blocks, all subjects were shown a fixation cross during the movement interval. The trials in which a fixation cross was shown in a given block were selected according to a pseudorandom schedule, which was the same for each subject. For the other half of trials, during the movement interval, subjects in the no feedback group saw a small still image of the first frame of the ear or toe demonstration video, while subjects in the feedback group saw a live video stream of their ear or foot relayed to the screen from the video camera. During training blocks, subjects in the no feedback group saw the still image of the model’s ear or foot during the movement interval of each trial, while subjects in the feedback group saw the live video stream of their own movements during the movement interval of each trial.
Subjects were instructed to watch the demonstration video closely, and to copy, as accurately as possible, the model’s movements in the movement interval. They were asked to copy the movements themselves as well as the timing of the model’s movements; that is, they were asked to perform five ear or toe movements during the movement interval, one approximately every 2000 ms.
Subjects’ ear and toe movement were video recorded throughout the experiment.
Video Scoring
There are 20 toe movement videos and 20 ear movement videos. In each video, there are three blocks, which were required to score. Each block lasts approximately 8 minutes. Each block consists of 20 mini blocks, each containing five trials. A score sheet was provided to record the scores.
Ear movement scoring:
- Count each movement made (a maximum of 100 in each block). Each movement should fall into the following categories:
- Ear only movement
- Ear plus eyebrow movement
- Ear plus face movement
- Ear plus whole scalp movement
- Whole head movement did not count.
Toe movement scoring:
- Count each movement made (a maximum of 100 in each block). Each movement should fall into the following categories:
- Sideways big toe movement only
- Sideways big toe movement, plus movement in other toes.
- Sideways big toe movement, plus whole foot movement.
- Sideways and upwards big toe movement
- Upwards big toe movement
The scoring of ear and toe movement both based on a 5-point scale in order to judge the size of each movement.
1=very small 2=small 3=medium 4=large 5=very large
Individuals who are blind to the experimental hypotheses have done video scoring.
Results
From the above, it is clear that the hypothesis has been carried and the results suggest that the following hypothesis is correct:
Under AIM: No difference between feedback and non-feedback condition for either toe or ear movements. Participants should learn in both.
Under ASL: More learning in feedback condition than non- feedback for ear movements (where there should be little or no learning).No difference between feedback conditioning for toe movements. Participants should learn in both.
Findings
It is seen that in the case study regarding opaque movements (Ear) for which the AIM (Active Intermodal model) has been used, the results have been depicted in Figure 1.
Similarly, the case study in which transparent movements (Foot) were depicted, for which the Associative sequence learning model (ASL) has been used is depicted in Figure 2.
Ear (Opaque movements) (AIM).
Foot (transparent movements) (ASL).
Upon application of the ANOVA test, it has been determined that the hypothesis has been proved correct:
- Under AIM: No difference between feedback and non-feedback condition for either toe or ear movements since the variance were not significant and therefore no major differences could be determined in this case.
- Under ASL: Under this method, it has been seen that in the case of ear movements, the differences have been significant which reinforces the fact that in this case, more learning in feedback condition than non-feedback has been found. However, in the case of feedback conditioning for toe movements, there have not been much of significant variances. Therefore, more learning in feedback condition than non- feedback for ear movements (where there should be little or no learning).No difference between feedback conditioning for toe movements. Participants should learn in both.
In conformity with the numerous research already carried out in the subject, it was found that the performance of the two control groups (Observers and respondents) was not vitiated by high rate of imitative responses, either unintended imitation or otherwise. The mean score was found to be less than significant ( 0.15 ) as against the overall score of 0.25. Thus the hypothesis could be seen to have been carried.
However, certain other observations also came out of the research study.
The results from this study support ASL Model circumspectly, but it is believed that are yet plenty of aspects to be discussed before a final decision is awarded.
It is seen that in a research of this sort, adults are the main respondents.
it is necessary that utmost care needs to be exercised in the method and mode of data collection, the actual experiments, various physical implements and accessories that are required for the study and also aspects of the safety and welfare of the children, who are undergoing the research survey and how it could be biased, or misjudged in its final assessment.
Next, it is seen that observers who intend to imitate the movements and observations giving rise to mutual activation in manner as described by ASL. This activation is normally inhibited so that it does not result in overt imitative actions. The observer’s intention is to be immobile or inflexible, or to perform a non-imitative movement. It has been noticed that in the case of adults, healthy human being, they may involuntarily perform certain special mannerisms, or nervous twitches, which may be performed for a longer period than it was intended.
But research studies have indicated that that these actions may not be spontaneously performed when they were in a position to observe themselves doing these actions, This how may not be the case as in the case of mental patients whose restrictive motor activities have been mal-functioning due to onset of frontal mental disease.
It is seen that the two theories, Active intermodal mapping modal (AIM) and Associated Sequence Learning Model (ASL) have been the mainstay upon which the aspects of imitation have hinged. These theories have made significant headway into the understanding of the conceptual knowledge and empirical studies that have gone into the framework of conceptual and practical knowledge of the imitation theory in its absolute terms. Although there have been other theories in vogue relating to imitation theory, by far, it has been AIM and ASL that have held sway in research matters. It is necessary to take a fresh look at these two aspects, and how these theories have helped in the cause of understanding them in its perspective.
What is important is the manifestation of imitation in the behavior pattern of people, not in the sense of learning but in the sense of imitation performance. The practical application of imitation in the empirical sense of the term, it is necessary to focus on imitation at the level of individual behavior, responses to external and internal stimuli, and how it could be controlled and improved,
The establishment of moderating in attitudes and behavior patterns involving imitations could be established and how the greater interests of society vis-à-vis the individual could be made. It needs to be seen how the psychological and neuropsychic aspects of imitating could be analyzed, researched in its fullest depth by competent professionals, and necessary inferences drawn.
Research on imitation: It is seen that imitation learning occurs mostly by chance. These are however, reinforced by observers when activities spread, but it is seem that on most occasions the imitating is spontaneous and the child does not wish to imitate the parent intentionally. Thus, for understanding cognitive approaches, it is not only necessary to refer to one’s body movements, but see it in relationships with other bodies and its activities. (Meltzoff, & Prinz, 2002).
Then only, the correct understanding of the concept of imitating can be gained and understood in its proper perspective.
Reference
Hoffman, Matthew W et al. (2006). Elsevier: A probabilistic Model of Gaze imitation and stared attention. Neural Network. P. 299. Web.
Alissandrakis, A., Nehaniv, C.L & Dautenhahn, K. (2008). Imitation with Alice: learning to imitate corresponding action across dissimilar embodiments. IEEE Xplore Web.
Brass and, Marcel & Heyes, Cecilia. (2003). Imitation: Is cognitive neuroscience solving the correspondence problem? Science Direct. Web.
Meltzoff, Andrew N & Prinz, Wolfgang. (2002). The imitative mind: Development, evolution and brain bases, Questia School. Web.
Mjølne, Asgaut. (2005). Towards emotional humanoid sociable robots: Major Assignment Applications of A1, Web.
Bird, Geoffrey., & Heyes, Cecila (2004). Imitation: Thoughts about theories: 1.3 Associate sequence learning, P.4. Web.
Brass, Marcel., & Heyes, Cecilia. (2005). Imitation: Is cognitive neuroscience solving the correspondence problem?, Elsevier. Vol.9. No.10. Web.
Yando, Regina., Seitz, Victoria., & Zigler, Edward. (1978). Imitations: A developmental perspective: Comparisons among imitation: theories on major issues, questia. P.48. Web.