An experiment to find out if seeing the eyes of a well known persons face is a factor of face recognition

Nicole Aves

An experiment to find out if seeing the eyes of a well known (celebrity’s) face is a factor of face recognition.

Abstract;

The aim of this experiment is to find out if participants can recognise the faces of well known celebrities if the celebrities’ eyes are blacked out and if the eyes of a person’s face are a major factor of face recognition. I predict that the participants will find it easier to recognise the celebrities’ faces in the condition where the eyes are not blacked out more than when the eyes are blacked out.

I used a lab experiment and independent group design. The target population from which my participants were used were the 20 students in my AS psychology class (opportunity sample). I randomly separated the 20 participants into 2 groups of 10 for each of the 2 conditions. I then delivered a brief and handed out the pictures of the celebrities face down in front of the participants. I then timed the participants for 2 and a half minutes while they turned over the sheets and wrote their answers upon it.

In consideration of my results I reject my null hypothesis and accept my hypothesis that the participants will find it easier to recognise the celebrities’ faces in the condition where the eyes are not blacked out. The results indicate that the eyes of a person’s face are an important factor of face recognition.

Introduction / Background:

Being able to recognise other people is of great significance in our lives. You can imagine how difficult life would be if you didn’t recognise your mother, brother, best friend or partner. Human beings seem to have an amazing ability to recognise faces. Standing (1973) showed participants 10,000 faces over five days. When they were shown pairs of faces, one of which they had previously been shown, together with a new one, they were able to identify the face they had seen 98% of the time. This ability is also impressive when people are asked to identify faces that they have not seen for some time, as demonstrated by the study by Bahrick et al (1975). Several years after their graduation from high school, Bahrick showed participants five pictures from yearbooks. One picture of the five was of a person with whom they had graduated. The results showed that the participants were accurate in their identifying 90% of their classmates. Even those participants who had graduated more than 40 years previously could identify 75% of their classmates correctly. This shows that recognition of faces shows little evidence of trace decay and therefore supports the idea that an individual can recall faces accurately for potentially a lifetime.

Aim;

In my study I will aim to find out if participants can recognise the faces of well known celebrities if the celebrities’ eyes are blacked out and if the eyes of a person’s face are a major factor of face recognition.

Hypothesis;

I predict that the participants will find it easier to recognise the celebrities’ faces in the condition where the eyes are not blacked out. My null hypothesis is that there will be no significant difference in face recognition between the two conditions (the celebrities’ eyes blacked out, and the celebrities’ eyes not blacked out). ...

This is a preview of the whole essay

Aim;

Hypothesis;

I predict that the participants will find it easier to recognise the celebrities’ faces in the condition where the eyes are not blacked out. My null hypothesis is that there will be no significant difference in face recognition between the two conditions (the celebrities’ eyes blacked out, and the celebrities’ eyes not blacked out). Any results will be due to chance.

Variables;

The independent variable is the variable that is altered or manipulated by the researcher. In this experiment the independent variable is whether or not the eyes of the celebrities are blacked out or not.

The dependant variable is the variable that is measured to see if the independent variable has had an effect. In this experiment the dependant variable is how many of the faces are recognised by each of the participants in both conditions.

A confounding variable is a variable that could affect the dependant variable but is not the independent variable that the experimenter is interested in. One possible confounding variable in my experiment is noise in the corridor outside the lab that the study was conducted (eg pupils outside the room talking). To control this extraneous variable I will place a sign outside the lab to let everyone within the proximity know there is a psychology experiment taking place and that silence is required.[1]

Method;

Design:

The method that I used was a lab experiment. I chose this method because lab experiments give the researcher the ability to place all aspects of the environment under control as far as possible, and is arranged specifically for that particular study. At the start of the practical section of my study I placed a sign outside of the room letting all people walking past that there was a psychology experiment happening, and to be as quiet as possible so as not to disturb the participants. This can be found in the appendix [1]. I also made sure that all of the participants were not sitting too close to each other and to perform the task in silence.

In this experiment I used independent group design to eliminate any demand characteristics (for example boredom or fatigue). I randomly selected the participants into the two conditions by allowing them to pick a folded piece of paper out of a plastic wallet that had either group1 or group2 written on it.

Participants:

The target population from which my participants were used were the 20 students in my AS psychology class. I randomly separated the 20 participants into 2 groups of 10 for each of the 2 conditions. The sample that I used was opportunity sampling as the experiment was based on the participant’s availability and willingness to take part. I also used random sampling to allocate the participants into the two conditions because then each of the participants would have an equal chance of being in either group.

Materials/Apparatus:

To each participant I gave a handout containing 20 pictures of 20 different celebrities, to one group however the eyes are blacked out (the same celebrities were depicted in both conditions). I did this to see if the participants’ face recognition ability varied if the eyes of the face could not be seen. There are examples of the handouts in my appendix [2]. I only marked the answers correct if the participants wrote the full name (or in some cases a nickname or well-known character name) of the celebrity.

Procedure:

This experiment took place on 21st January 2005 in lab 5 at Barton court grammar school in Canterbury. After randomly separating the participants into the 2 groups (described in the participants section of the method) I asked group 2 to leave the room. Once they had left I delivered my brief (found in the appendix [3]) to fully inform the participants of how the experiment was to take place, give them the right to withdraw and ensure them of confidentiality and anonymity to avoid any ethical issues. I then handed out the pictures of the celebrities (with visible eyes) face down in front of the participants. I then timed the participants for 2 and a half minutes while they turned over the sheets and wrote their answers upon it. At the end of the time I asked them to stop writing so I could collect them in. then to finish I delivered my debrief outlining the aim of the study (also found in the appendix [3]), thanked them for taking part and asked them to leave the room. I repeated the same procedure with group 2. This time with the pictures of the celebrities with blacked out eyes.

Results:

A table to show the difference in face recognition between celebrities faces with eyes and faces with the eyes blacked out.

Raw data can be found in the appendix [4]

The table shows that the participants in condition 1 recognised more of the celebrity’s faces on average than those in condition 2.

Discussion;

In my study I aimed to find out if participants can recognise the faces of well known celebrities if the celebrities’ eyes are blacked out and if the eyes of a person’s face are a major factor of face recognition. The participants in condition 1 (celebrities with eyes) recognised the faces easier than the participants in condition 2 (celebrities with their eyes blacked out), as shown in the results the mean score for condition 1 is 17/20 and 13/20 in condition 2 – a difference of 5/20 in favour of condition 1. I chose to use the faces of well-known celebrities so that they could be recognised by all of the participants.

The results of my experiment could support both of my background studies. Bahrick et al (1975) found that participants were accurate in their identifying 90% of their classmates and 75% after 40 years. Standing (1973) found that participants could identify faces from a huge selection (10,000) 98% of the time. My results show that my participants could identify 85% of the known faces (condition 1) accurately. The participants will have a lot of faces stored in their long term memory, so to be able to recognise one face and match it with one of the many in their memory accurately supports Standing. It is quite possible that the participants haven’t seen the faces of some of the celebrities very recently but still recognised them correctly, this therefore supports the study by Bahrick et al.

The sample of people tested could be considered a limitation of my experiment. I only used an opportunity sample of 20 people from the same subject (psychology) in the same school. This is not representative of the target population. The results of my experiment cannot be generalised, as the target population is not big enough or varied enough. if I was to further my research into face recognition I could rectify this limitation by increasing the number of participants and select them from various schools. For example 200 participants from 4 different schools.

Conclusion:

References:

Erika Cox 2002, AS level psychology for AQA specification B, Oxford University press.
2002, AS-level psychology revision guide, coordination group publications ltd

Appendix [3]

Brief:

I would like to take this opportunity to thank you all for taking part in my AS level Psychology coursework. All your results will be kept completely confidential, as I will only refer to you by number. You all have the right to withdraw from this experiment at any time and take no further part in this investigation, and naturally your results will not be used. I will give you each two sheets with twenty celebrities’ enpicted. All that is required of you is to write down by the picture the name of as many of the celebrities as you can. I will give you two and a half minutes to complete the task.

Debrief

The aim of the investigation was to see if being able to see the eyes of a known person’s face is an important factor of face recognition. I would like to remind you that I will keep your results confidential, but if for any reason you would like to withdraw your results you are entitled to do so. I expect that all your results are within the normal range, and if you would like to know your results then you may see me at a later date. Thank you again for taking part and well done.