Evaluating priming effect on perception of ambiguous figures PDF

Preview

Summary

Evaluating priming effect on perception of ambiguous figures...

Description

Quantitative Research Methods of Psychology Evaluating Priming Effect On Visual Perception Of Ambiguous Figures Word Count: 1929

Abstract: Ambiguous figures are defined as images that can have multiple interpretations even though their physical properties do not change. Although it is not yet proven how perception is affected, the top-down and bottom-up theory suggests that visual perception is based on the information received and on previous experiences to form a context of what is being perceived. The purpose of this research is to evaluate the priming effect on visual perception of ambiguous figures, assessed by an online survey-based experiment. 427 participants with no visual impairments took part on this experiment and were allocated automatically to either the priming condition group or the control group. The results of the experiment were significant and it suggested that target image recognition score is higher on the priming condition group than the scores on the control group. Implications and limitations are discussed and further research is suggested.

Introduction: An ambiguous figure is a visual illusion in which an individual can switch from one visual image to another equally valid one. Wimmer, Doherty and Collins (2011)

defined

ambiguous

figures

as

images

that

can

have

multiple

interpretations even though their physical properties do not change. Palmer (1975) indicates that it is not proven how context may affect perception, but

appropriate

context

supports

correct

identification

of

objects

and

inappropriate context should delay its recognition. Possible explanations for the perception of two or more images of ambiguous figures fall on the bottom-up and top-down perceptual set (Intaite, Noreika, Šoliunas & Falter, 2013). The bottomup theory supports that visual perception is based only on the incoming information, while the top-down processing uses previous experiences and expectations to form an image of what is being perceived, and reversals of these images arise from adaptation and recovery (Intaite et al, 2013). Due to specific processes, our brain is capable of perceiving one image first and then a second one of an ambiguous figure (Wimmer et al, 2011). For instance, with the ambiguous figure ‘Rubin Vase’, it is possible to see first either a white vase in a black background, or silhouettes of two opposing faces on a white background (see Figure 1 below). After it is acknowledged that is possible to perceive both images, the brain can reverse them to see either image or both at the same time.

Figure 1. Ambiguous image ‘Rubin Vase’. Retrieved on November 11th, 2019 from Google search. A study conducted by Balcetis and Dale (2007) supports that the perception of an ambiguous image can be biased when an individual is primed with written

material prior to seeing an ambiguous figure. The participants of this experiment (n=58) were undergraduates from Cornell University, where one group (n=32) were shown a paragraph related to automobiles and another group (n=26) read a paragraph regarding astronauts and space exploration, then both groups were shown a series of images unrelated to both texts and asked to write a description of them, after this, participants were shown the target image and asked to describe it (Balcetis & Dale, 2007). See figure 2 for target image. This demonstrates that visual perception can be manipulated, not only with images, but also with information obtained through other means.

Figure 2. “Photograph that includes a sign/moon ambiguous object”. Reprinted from “Conceptual set as a top-down constraint on visual object identification”, by E. Balcetis and R. Dale, 2007, Perception, 36, p. 585. The findings of Goolkasian and Woodberry (2010) also support this conclusion. An experiment conducted with students, where either an image or word format prime was shown as one of the interpretations of the ambiguous figure was heavily altered and details of the other interpretation were reduced, found that participants were able to identify the images despite being modified (Goolkasian & Woodberry, 2010). The literature above suggests that priming does have an effect upon visual perception of ambiguous figures, with these studies conducted in controlled environments, as opposed to the present experiment. This study aims to

evaluate the priming effect on visual perception of ambiguous figures assessed by an online survey-based study. Thus, the alternative one-tailed hypothesis of this experiment is that target image recognition score will be higher on the priming condition group than the scores on the control group. Therefore, the null hypothesis is that target image recognition score will not be higher on the priming condition group than the scores on the control group.

Methods: Design: This experiment was designed to evaluate whether priming has an effect on the visual perception of ambiguous figures. In order to analyse this, a between participants experiment was conducted with a priming condition group and a control group. The independent variable is the priming condition and the dependent variable is the target image.

Participants: There was a total of 489 participants who took part on this experiment by opening the link and 62 of them did not complete the entire study. 210 participants were part of the control group, where the maximum age was 72, the minimum age was 18 and 6 of the participants did not provide their age. The mean age on the control group was 30.42. There were 217 participants on the priming condition group, where the maximum age was 83 and the minimum was 17, and 5 of the participants did not give their age. The mean age for the priming condition group was 30.39. On the control group, there were 140 participants who identified themselves as females, 65 as male and 1 of them as other (e.g. non-binary). There were 4 participants on the control group that did not provide their gender, while on the priming condition group, there were 2 participants that did not give their gender, 134 participants were female, 80 were male and 1 identified as other.

Materials: Participants accessed the experiment through an online link provided by the experimenters. The survey was comprised by 10 trials. On the priming condition group, the priming image was shown for 5.5 seconds, followed by the target image, while on the control group, a control image was presented for the same amount of time, followed by the target image. On both groups, participants got a point for every time they saw the target image first, they did not get a point if they see the other image first or did not see either. See Figure 3 for a sample of trial. Minimum possible score was 0 and maximum score was 10. A higher score means participants saw the target image first, therefore, it would mean that priming does have an effect on the visual perception of ambiguous figures. Images were retrieved from a Google image search. See instructions on appendix A.

Priming Condition Group

Control Group

Figure 3. Sample of trial on priming condition group and control group with the same ambiguous figure duck/rabbit. Images were taken from a Google Image search.

Procedure: An online link was used to access the survey, which was provided by the students of UWL to participants, who were general population of a minimum age of 18. Individuals with visual impairment and Psychology students at UWL were advised not to take part on the experiment. This experiment was conducted anonymously, and only age and gender were asked, which could be left blank. Participants could withdraw from the experiment at any moment by closing the browser with the survey. Ethical approval was granted by the Psychology Ethics panel at UWL, there were no perceived risks with this experiment. Participants were briefed with an information sheet and consent form before being presented with the survey, as well as given the example of an ambiguous figure. For the purpose of the study, participants were not aware that they were being primed. Participants were randomly allocated to either the priming condition group or the control group by the Qualtrics Survey Solution system, which was used to conduct the online experiment. During the survey, participants in both groups were shown an image for 5.5 seconds, followed by the target image, then given the option to choose which image of the ambiguous figure they saw first, or if they did not see any image of the possible ones. Lastly,

participants were debriefed and informed in which group they were. It was explained in more depth the purpose of the study and, on the cases of the priming condition group, it was disclosed that they were being primed.

Results: Previous research suggests that the visual perception of ambiguous figures can be purposely biased by a stimulus shown before the figure. This study aims to evaluate the priming effect on visual perception of ambiguous figures assessed by an online survey. Data was collected and processed on SPSS statistics version 25, the results are presented in the following table:

Table 1: Difference between scores on the control group and the priming condition group Control Group Prime Group

Mean 4.52 4.98

Standard Deviation 1.20 1.30

The mean score for the control group was 4.52, whereas the mean score for the priming condition group was 4.98. The standard deviation for the control group was 1.20 and for the prime group was 1.30. The target image recognition score on the priming condition group was higher than the overall score on the control group. The experiment was conducted between participants. The data collected on the dependent variable was ratio and the Kolmogorov Smirnov test suggested the data was not normally distributed. However, the number of participants was greater than 200, therefore, the parametric unrelated (or independent) t-test was run. Results were significant, and it showed that the priming condition group scored higher than the control group on the perception of the target image. Levine’s test confirm that the data has equal variances and the unrelated t-test verifies that t (425) = 3.75 , p < .001. This is also illustrated by the figure below:

Figure 4: Chart bar showing difference on target image recognition score by group. Discussion: The present survey-based experiment has been conducted to evaluate the priming effect on visual perception of ambiguous figures. The results of this experiment have supported the alternative hypothesis stating that target image recognition score will be higher on the priming condition group than the scores on the control group. These results are consistent with the conclusions of Balcetis and Dale (2007), who investigated the effect of priming on visual perception of ambiguous images by

language-induced

cues,

using

a

within

participants

experiment.

Nevertheless, as opposed to the present experiment, the study conducted by Balcetis and Dale (2007) does not report descriptive statistics for age and gender and does not state the field of education of participants, visual impairments or whether participants have interacted with ambiguous figures before the experiment was conducted. In contrast, participants of the experiment conducted by Googlkasian and Woodberry (2010) were psychology students from the University of North Carolina, mostly females (73%) and with no visual impairments. Additionally, the method differs to the present study, as the experiment of Goolkasian and Woodberry (2010) consisted of 72 trials, where participants used a 6-point numerical rating scale, of which 36 were heavily altered. Participants were

primed with either a word or image format, with the prime appearing for 5 seconds followed by a modified version of the ambiguous image. It is worth noting that the main difference between the studies discussed and the present experiment is that the latter was conducted using an online system that allowed participants to complete the survey from their location, avoiding the need to travel to a specific site, which could also implicate a limitation, as there is a possibility that participants got distracted while completing the survey, or were completing it with the help of someone else, consequently, affecting the overall score and the initial perception of the target image. The literature discussed and the present study support that written and visual prime does have an effect on visual perception of ambiguous figures. Nonetheless, this research has been conducted including male and female participants on both priming condition and control groups, not considering the possibility that perception may differ between genders or age groups. On this matter, future research is needed to evaluate whether gender and age affects visual perception on ambiguous figures.

Further investigation is suggested to analyse the possibility that senses, such as hearing and smell, may also be used as prime to evaluate the effect on ambiguous figures. The use of senses as priming may also be of use for patients with Amnesia, Dementia or Alzheimer, as it might strengthen memory skills by task repetitions.

References: Balcetis, E. & Dale, R. (2010). Conceptual set as a top-down constraint on visual object

identification. Perception, (36), 581, 595.

Goolkasian, P. & Woodberry, C. (2010). Priming effects with ambiguous figures. Attention,

Perception, & Psychophysics, 72(1), 168-178.

Intaite, M., Noreika, V., Šoliunas, A. & Falter, C. (2013). Interaction of bottom-up and

top-down processes in the perception of ambiguous figures. Vision

Research, (39),

24-31.

Palmer, S. (1975). The effects of contextual scenes on the identification of objects. Memory

& Cognition, 3(5),519·526

Wimmer, M. C., Doherty, M. J. & Collins, W. A. (2011). The development of ambiguous

figure perception. Monographs of the Society for Research in Child

Development,

76(1), 1-130.

Appendix A: Participant instructions

This study is looking at ambiguous figures. An ambiguous figure is a picture that contains more than one different image within it. You will be asked what image you saw first. Don't worry if this isn't quite clear to you, as you will be given an example before you start. Remember that you can withdraw from the study at any time, just by closing down your browser (click the little x in the top right hand corner of the screen).

You will be asked your age and gender (you can always leave these blank if you prefer) and then given the example ambiguous figure before you begin.

Please click the blue arrows in the bottom right hand corner of your screen to continue.

Please indicate your gender: ________________________________________________________________

Please indicate your age in years: ________________________________________________________________...