Different Character Attributes and Their Effect on Forming Impressions

Different Character Attributes and Their Effect on Forming Impressions


The purpose of this experiment was to discover whether presence of the words “warm” or “cold” have an influence on participants’ first impressions of people. Participants rated their own personal attributes and the attributes of a fictitious person. Based on the condition group, participants were told that the person was either “warm,” “cold,” or neither “warm” nor “cold.” Ratings of the person were affected by participants’ condition group. Effects were significant only with regards to ratings social traits of the person. Perceived liking of the person was significantly affected by participant’s condition group. And, participants gave a higher rating for liking when there was perceived similarity with the person. These results were consistent with findings of previous studies by Asch and Kelley.

Different Character Attributes and Their Effect on Forming Impressions

Our natural instinct as human beings is to form impressions of people when we meet them. We cannot help but make judgments of our liking or disliking of a person based on details we know about that person. It is extremely difficult not to form opinions about another person if we know characteristics about that person. The following study was conducted in attempt to examine how people form first impressions on someone based on adjectives describing that person. Past studies have shown that the presence or absence of certain words describing the person, such as the words “warm” and “cold,” can drastically change a person’s impression of that person. These words would be examples of central qualities. Secondary qualities would be other attributes, such as “industrious,” “critical,” “practical,” and “determined.” It has been argued that nature of a person’s central qualities changes how their secondary qualities are perceived.

S. E. Asch and H. H. Kelley both did previous studies relating to this topic. One of the goals of the current study is to validate both Asch’s and Kelley’s findings. Although this study is very similar to their studies, there are some differences between them. For instance, in Asch’s study, he read the qualities aloud instead of having them written on paper. Also, our study did not use as many attributes as Asch used in his study.
The hypotheses in this study were formed based on the findings of Asch and Kelley. There were four hypotheses in this study. The first hypothesis was that people’s knowledge of someone’s central traits would affect their general rating of that person. This relates to Asch’s study in which he found that certain labels given to a person could completely alter someone’s impression of the person. The second hypothesis was that the effect in the first hypothesis would only be true for ratings of social traits. The third hypothesis was that the degree to which a person likes another would depend upon the condition group they are in. In other words, it was predicted that participants in the “warm” condition would give higher ratings for liking than participants in the “neutral” and “cold” conditions would.

The fourth hypothesis was that the more people perceive similarity to another person, the higher their ratings of liking will be. The reasoning behind this hypothesis comes in part from Byrne’s study in which he discusses how similarity between people leads to attraction or liking between them. Therefore, it would be expected that if people think that they are similar to another person, they would tend to like that person more so.


Participants - One hundred and fifty college-age youths served as participants in this experiment. Their ages ranged from 20 to 22, with both males and females participating. One third of the participants were in the “warm” condition, one third were in the “neutral” condition, and one third were in the “cold” condition. Everyone in the experiment was participating as a class exercise in an undergraduate psychology class at The University of North Carolina at Chapel Hill. None of the participants were paid for participation in this study.


All participants were given questionnaires. The questionnaires were made up of two sections. First, the questionnaires asked participants to rate themselves on a scale from one to seven on a variety of different characteristics. The characteristics were: helpful/unhelpful, determined/wavering, moody/stable, frivolous/earnest, submissive/unyielding, pessimistic/optimistic, humorous/humorless, serious/unserious, imaginative/unimaginative, tolerant/intolerant, vain/humble, naïve/experienced, clumsy/skilled, good-natured/mean-spirited, persistent/inconsistent, and boring/lively.
The second section of the questionnaire involved rating a fictitious character named “Alex.” Depending on the condition group of the participants (“warm,” “neutral,” or “cold” variables), Alex is described differently. The description of Alex was the following: “Alex is 25 years old, a college graduate, and is unmarried. People describe Alex as an industrious person, critical, {warm, cold, or word was omitted}, practical, and determined.” After participants read the description of Alex, they were then asked to rate Alex on the same characteristics as they rated themselves in the first part of the questionnaire. In addition, participants were asked to rate how much they would expect to like Alex on a scale from one to seven.


Participants were divided into three conditions, based on whether they were in the “warm,” “cold,” or “neutral” group. Participants were unaware that they were divided into different groups. All participants were first asked to complete a questionnaire in which they rated themselves of their own characteristics. They then gave ratings of Alex. The last question asked how much the participant thinks he/she would like Alex. The data was then recorded and analyzed.


In this study, a 3x2 repeated measures analysis of variance was conducted. All four hypotheses were supported through the data. It was found that there was a significant effect in Alex’s ratings based on the warm-neutral-cold variable. The first hypothesis was supported, based on the findings of this study. It was found that participants’ knowledge of Alex’s central trait significantly effected their general ratings of Alex (F (1,147) = 87.04, p < .05). The second hypothesis was also supported with the data in this study (F (2,147) = 7.49, p < .05). This suggests that the effect was only significant in respect to Alex’s social traits, not his intellectual traits.

The third hypothesis was also supported, which suggested that the degree to which participants rated high liking of Alex was determined by the condition group they were in. More specifically, participants in the “warm” condition liked Alex the most. Liking of Alex was less for participants in the “neutral” condition, and liking was least for participants in the “cold” condition. Three t-tests were conducted to determine whether the conditions were significantly different from each other. The “warm” and “neutral” groups were not significantly different from each other (t(1) = 1.80, p < .05). However, the “neutral” and “cold” groups were significantly different from each other (t(1) = 4.08, p > .05). Likewise, the “warm” and “cold” groups were also significantly different from each other (t(1) = 5.79, p > .05).

For the fourth hypothesis, a regression analysis was performed (R = .52, p < .05). This suggests that the participants’ perceived similarity was related to his/her perceived liking of Alex. It was also found that the more social similarity between participants and Alex, the more participants gave higher ratings for liking. This suggests that social similarity is more important than intellectual similarity in determining liking.


All four hypotheses were supported in this study. It was found that knowledge of Alex’s central traits (warm, neutral, cold) affected ratings of that person, in terms of social characteristics. It was also determined that the degree to which a person liked Alex depended on whether the participant was in the warm, neutral, or cold condition group. And, it was also found that the more similar a person was to Alex, the more the participant liked Alex. These results are consistent with the findings of Asch and Kelley. Therefore, it can be concluded that central traits, such as labeling a person “warm” or “cold,” do have an effect on forming first impressions of people.

One of the problems with this study is that it cannot be easily generalized to real life. This study lacks external validity. For instance, the character “Alex” was not a real person. Participants may not have formed the same impressions about Alex if he/she had been a real human being.
Also, because participants in this study read the characteristics of Alex (as opposed to having the characteristics being read aloud to them), they may not have noticed the central trait as much as a secondary trait. For example, a participant may have paid more attention to the word “critical” than the word “warm.” In this case, this participant would have skewed the data. All in all, this study was useful in confirming Asch’s and Kelley’s results. It is important to understand the implications of labels on forming impressions. Future research and more experiments on this topic may need to be conducted to better understand how people form impressions and make judgments of others.


Asch, S. E. (1946). Forming impressions of personality. Journal of abnormal and social psychology, 41, 258-290.

Byrne, D., & Nelson, D. (1965). Attraction as a linear function of proportion of positive reinforcements. Journal of personality and social psychology, 1, 659-663.

Kelley, H. H. (1950). Personal perception: the warm-cold variable in first impressions of persons. Journal of personality, 18, 431-439.