In this post
In this study, the researchers were interested in finding out whether or not the way in which someone perceives an image will affect the way that it is reproduced. It should be the case that if people are all looking at the same thing that they will reproduce it in the same way, but Carmichael et al. did not necessarily believe that this would be the case.
Aim
To see if words used to describe an object whose image was not clear could affect the way participants would later recreate the image in the form of a drawing.
Method
This was a laboratory experiment with an independent measures design. Thirty-five male participants and 60 female participants (95 altogether) took part and were divided into two groups, the experimental group with 86 participants and the control group with nine participants.
Participants were shown 12 images and, between each, blank images were shown but an introduction to what was coming next was made by the researcher: “the next figure resembles…” and the end of this would be filled by a word from either list one or list two for the experimental condition participants. The control participants did not receive any such verbal cue.
Once all participants had seen all 12 images, they were asked to reproduce them in any order. If necessary, the sequence of images (and verbal cues for the experimental condition) was repeated until all participants had a drawing for every image that they had seen.
Drawings were then assessed as to their likeness of the original image that had been presented in terms of:
- Being almost identical
- Being different in line lengths or curves
- Being different in proportions of parts of the image
- Having omissions (things missed out) or additions
- Being almost entirely different
Ratings were carried out by two different researchers and when an agreement could not be made in terms of the factors above, a third researcher was brought in to help decide.
Only reproductions that fell into category 5 (as seen in the image below) were used and compared to the word list (the word list being the dependent variable).
Participants were also asked how they believe they had performed in the task and their answers were noted down.
Results
- Participants who had heard a word tended to have created reproductions that looked like those words
- More than 3,000 reproductions were assessed and 905 of these fitted into category 5
- 73% of participants who had words from list 1 produced drawings which resembled the spoken word
- 74% of participants who had words from list 2 produced drawings which resembled the spoken word
- 45% of participants from the control group produced drawings which resembled images from list 1 or 2.
Conclusion
The researchers concluded that hearing a word spoken affected the participants’ reproduction of the image, and therefore memory must be reconstructed. It was also concluded that the verbal context in which images are perceived must affect later reproduction because, reproduced, the memory of the word changes the way in which the drawing is represented.
Strengths of the study
- Use of a control group enabled the researchers to know that the drawings that were reproduced were not distorted because the original images were ambiguous
- Use of two separate lists helps to confirm the role of verbal labelling in distorting recall.
Weaknesses of the study
- The research took place in a laboratory and therefore lacks ecological validity
- The sample size of 95 is low, as is the amount of people who were in the control group (9) making it difficult to generalise the findings to a larger population
- The test was completely artificial, as no one would be asked to do this as part of real-life perception and recall.
