It has been suggested that a configural model exists concerning the formation of a person's impression of another in which certain central traits may have a disproportionate role. This theory was put forward by Asch (1946) in a series of investigations that saw him varying one trait within a set of traits, and then gaining a record of the final impressions that were made by subjects after hearing these traits. An analysis was then made addressing whether or not these varied traits had any effect on the final impressions. Apparently Asch was of the opinion that the final impressions would vary according to the presence of a different central trait. This would suggest that all traits that are given to the subject are combined into a single image, or Gestalt, of the hypothetical person in question. Asch's experiment was also further validated by a more 'naturalistic' replication by Kelley (1950), who introduced a guest lecturer to students, describing the newcomer by listing a series of traits. He varied the 'central trait' however, in the same manner as Asch did in his experiment, that is, by swapping a single trait within the set with another trait - 'cold' became 'warm'. Kelley then asked the students to evaluate the lecturer, and received different impressions according to which central trait was given to the students present at a particular lecture. Again, this supports the Gestalt view that impressions are integrated wholes based on central traits. Asch also suggested the presence of peripheral traits, being traits which have a rather insignificant role on the configuration of the final impression. To demonstrate this Asch inserted polite or blunt into the position previously occupied by warm or cold. The final inferences made about the 'polite' individual were not radically different from those formed concerning the 'blunt' person, leading Asch to the hypothesis that particular words within a given context did not affect the final impression to any great degree.

Another theory is the presence of the 'halo effect', best demonstrated in studies by Dion, Berscheid and Walster (1972). The halo effect (and it's opposite, the 'negative halo' or 'fork-tailed' effect) suggests that a person can be labeled as 'good' by the presence of one particular trait (usually attractiveness) and then all good qualities are attributed to that person, and bad qualities ignored or modified to concur with this impression. Although this seems to support Asch's Central Trait theory, it works to the detriment of a Gestalt based, rationally coherent impression as it 'glorifies' one trait rather than a combined set.

Therefore, considering both approaches in this experiment, it is expected that a change in one trait dimension within a set of traits leads to a difference in the final impressions formed.

Method

Subjects

9 subjects with a mean age of 19, consisting of 6 females and 3 males, consented to be a part of this experiment after being requested to do so.

Materials

The list of words initially presented consisted of : intelligent, skilful, industrious, determined, practical and cautious; with either warm, cold, or polite being listed in addition between industrious and determined. The next list of six traits consisted of generous, wise, happy, goodnatured, reliable and important; and the subject made a rating of these words on a 1 to 7 scale; 1 being very unlike person X, 7 very much like person X and 4 neither.

Procedure

First, each participant is shown the Plain Language Statement, which introduces them to the task and then asked to give their informed consent by signing a copy of the Informed Consent Form.

Second, the participant is randomly assigned to either Group A (warm), Group B (cold) or Group C (polite) in such a manner that once all nine participants are assigned a category, each Group consists of three subjects (or more if further participants existed). The participant is then read the following instructions: "I am going to read you some words which describe Person X. Please listen carefully and form a general impression of Person X."

Next, a list of traits were presented to the participant with either warm, cold, or polite being inserted in the middle of intelligent, skilful, industrious, determined, practical and cautious - depending on the Group that the participant was assigned to. The following instructions were then read to the participant: "Now I am going to read a list of traits which may or may not describe Person X. When I read each trait, tell me the extent to which you think it describes Person X. Give a rating from 1 to 7, where 1 means that the trait is very unlike Person X, 4 means that the trait is neither like nor unlike Person X and 7 means that the trait is very much like Person X." The participant was then also presented with a copy of the scale, as presented in Table 1 (Appendix A).

The subject was then presented with the traits generous, wise, happy, goodnatured, reliable and important and their rating recorded on Table 2 (Appendix B). After their responses were recorded, the particiapnts were asked if they experienced any difficulty forming an impression, told the purpose of the experiment itself and then asked if they had any questions to ensure that they were fully debriefed and understood the experiment and their part in it.

The mean and standard deviations of the data given in Table 2 (Appendix B) was then calculated in order to determine the average response to each question and the range in which responses differed, shown in Table 3 below.

Results

An examination of Table 3 shows that little deviation existed between the scores given by each individual within a group. Each group had different means concerning the relevance of a particular trait to Person X, that is, it can be observed which group thought a trait was especially relevant to their impression of Person X in comparison with the other groups. It can also be noted that Group B (cold) gave considerably lower scores concerning what may be interpreted as positive traits, whereas Group A (warm) gave comparitively higher scores. Group C (polite) is similar to Group A (warm) in this respect.

Table 3. Mean scores for participants given the description containing 'warm'; for participants given the description containing 'cold'; and for participants given the description containing 'polite'. (Standard deviations in parentheses.)

Discussion

The results confirm the hypothesis that the final impressions will be different according to the variation of a single trait within the set of traits given, yet, evidence for both the existence of Asch's Central Trait Theory and the 'halo effect' are present. This would suggest that these are not in fact conflicting theories but analogous. That is, they both exist as recognisable phenomena within the complete process of impression formation.

The findings of Asch and Kelley are supported in that particularly strong impressions have been formed of Person X, dependant upon a variance of the 'central trait', demonstrated by the fact that similar responses were given by each individual in their respective grouping. And whereas one trait may have been considered particularly relevant to Person X by one group, another group deemed it less important, eg. Group A (warm) saw 'happy' as being applicable to Person X (mean - 6.33), whereas Group B (cold) considered 'happy' as being less appropriate (mean - 2.67).

It is also important to note that the responses given for 'polite' and the responses for 'warm' are somewhat similar in that they exist on the same side of the scale (Very Unlike or Very Much Like Person X) in comparison with the 'cold' impression of Person X that was formed by Group B. This suggests some similarity between warm and polite within the minds of the participants respectively, and/or it may suggest that one of these traits is a peripheral trait in that it has little influence in configuring the final impression.

The polarization of the results also gives credence to the presence of a halo effect. Considering that both the warm and polite groups were more likely to associate positive characteristics with Person X. This would suggest that the central traits had a dramatic effect on the positive or negative impression that was formed as the differences between the uniform words was rather ambivalent (the real peripheral traits in this study). And from these positive or negative impressions of Person X, a halo effect came into play. However, the presence of a halo effect cannot be proven as conclusively as the existence of Asch's Central Trait Theory, and so remains something of a hypothetical in this respect.

Further studies would therefore focus on isolating the presence of the halo effect, that is, trying to statistically prove its presence.

References

O'Shea, R. P. (1993). Writing for Psychology: An introductory guide for students. Sydney: Harcourt, Brace & Javanovich.
Schneider, D. J., Hastorf, A. H., & Ellsworth, P.C. (1979). Person perception and the Asch paradigm. Social Psychology A - Reader, Reading 2.1, Deakin University.
Sears, D. O., Freedman, J. L., & Peplau, L. A. (1985). Forming impressions. Social Psychology A - Reader, Reading 2.2, Deakin University.
Vaughan, G. & Hogg, M. (1995). Introduction to Social Psychology. Sydney: Prentice Hall.

PRACTICAL REPORT - APPENDIX A
INFORMED CONSENT FORM

DEAKIN UNIVERSITY
SCHOOL OF PSYCHOLOGY

I, __________________________________________________of
_________________________________________________________________________________________________________
Hereby consent to be a subject of a human research study to be undertaken by Daniel C. Kelly.

and I understand that the purpose of the research is to assess the impact of certain word descriptions on the impression that is formed of a hypothetical person. I understand that I will be read some words that describe a hypothetical person, that I will be asked to form an impression of that person and that I will be required to indicate the extent to which certain words describe that hypothetical person.

I acknowledge that

1. The aims, methods and anticipated benefits and possible hazards of the research have been explained to me.

2. That I voluntarily and freely give my consent to participate in such research study.

3. Individual results will not be released to any person.

4. That I am free to withdraw my consent at any time during the study, in which event my participation in the research study will immediately cease and any information obtained from me will not be used.

Signature: ____________________________________________

Date: _______________________________________________

PRACTICAL REPORT - APPENDIX B
PLAIN LANGUAGE STATEMENT

You are invited to participate in a simple study, conducted as part of my studies in Social Psychology at Deakin University. The aim of the study is to assess the impact of certain word descriptions on the impression that is formed of a hypothetical person. You will be read a list of words and asked to form an impression of a hypothetical person. Then you will be asked to rate the similarity of the person to certain descriptions or traits. Finally, you will be asked some questions about your experience of the task.

The study will take approximately 10-15 minutes of your time. You should NOT write your name on any handouts. The results will be detailed in an assignment, submitted for assessment. However, I will ensure that your name will be removed from the data. You are free to withdraw from the study at any time, please let me know if you are unhappy about any aspect of this study or you can contact The Chair, HPS204 Social Psychology A, School of Psychology, Deakin University, Geelong. Victoria. Australia.

Signature (of student conducting the study)

__________________________________________