Face Recognition Theory Assignment

1

2
Abstract
This study explores the theory of whether we are more accurate at recognising
names or occupations of a person from just looking at their face. With their face
recognition theory in 1986, Bruce and Young suggested that we can recall the
name, occupation, and other sorts of information about a familiar person when
looking at their face. To this end, students were recruited if they studied
Psychology at University of Bedfordshire Luton campus. The study data was
collated over three years. They all were on the MSc psychology conversion
program (N=190). The participants consisted of 27 male (14.2%), 160 female
(84.2%) and three participants had their gender missing. Participants were aged
between 19 and 57 years (M = 27.5, SD = 8.89). Participants were asked to
complete a survey using the Qualtrics survey software. The survey consisted of
an exposure and test phase. The exposure phase consisted of timed picture
slides of 30 faces from the Karolinska Directed Emotional Faces database with
name and occupation attached. All single faces had a unique name and
occupation. The test phase had ten unchanged slides from the exposure phase,
ten with different names and ten with different occupations. Participants were
asked to recall and respond by choosing between if there was no change, a
change in name or a change in occupation. A paired-samples t-test indicated that
scores were significantly higher for those that identified a change in occupation
subscale (M = 7.46, SD = 2.47) than for the subscale of those who identified a
change in name (M = 4.47, SD = 4.47), t(189) = -14.69, p < .001, d = 1.29.

3
Face Recognition and Bruce & Young 1986 theoretical framework
Face recognition is the ability to identify an individual by looking at their
face. Recognising a face provides a person's identity and the essential
information for social interaction such as name, age, gender, occupation, mood,
and emotions. Due to the complexity of face recognition, extensive research has
been done to gain a better understanding. (Goldstein & James, 1983) stated that
"The face is the most important visual stimulus in our lives probably from the first
few hours after birth, definitely after the first few weeks". The ability to recognise
faces is essential in every person's life differs significantly from object
recognition(Bruce & Young, 1986).
Bruce and Young (1986) theoretical framework for face recognition has
been the most influential. The framework suggests that eight components make
up the face recognition process.
These components are:
 Structural encoding: this is when the person can observe, recall another
person and assign them names or attributes.
 Expression analysis: this process allows the person to pick on emotional
cues from facial expression.
 Facial speech analysis: it uses facial motions to assist in speech
perception.
 Direct visual processing is the ability of the processing any information
received through the eyes.

4
 Face recognition nodes: the process of using structural features to identify
a person.
 Person identity nodes: these provide stored information of individuals such
as occupation and interests.
 Name generation is the process of recollecting or assigning an individual's
name.
 Cognitive system: this holds additional information and assigns which
other components receive attention.
The Bruce and Young framework suggest that the processes of identifying
familiar and unfamiliar faces involve different components. Familiar face
recognition involves structural encoding, face recognition units, person identity
nodes and name recognition. On the other hand, unfamiliar face recognition
involves structural encoding, expression analysis, facial speech analysis, and
direct visual processing (Severin et al., 2005).
Bruce and Young's model focused more on familiar faces. Bruce and Young
(1986) proposed that only the appropriate person identity node would give
access when generating a name. On the assumption that individuals have no
brain damage, they should put names to faces without knowing anything else
about the other person. However, some patients demonstrated a different pattern
disproving that theory. They also proposed that name recall was more
challenging to achieve than other information like an occupation. Names are
abstract and usually arbitrary than other types of identity-specific information
such as occupation, hobbies and other things. Researchers have also conducted

5
experiments such as (McCluney & Krauter, 1997; Terry, 1994) with similar
outcomes.
This study also aims to prove this hypothesis that name recollection from face
recognition stimuli is more challenging to achieve than occupation recollection

6
Method
Design:
The institutional review board (IRB) approved the following research
design. The study is a repeated measures quantitative study using both
descriptive and inferential analysis to analyse the data. Data collected were the
age, gender, and the accuracy of identified changes in name and occupation
from pictures provided. The type of information provided on the pictures, which
are the name versus the individuals' occupation in the picture is the independent
variables. Simultaneously, the accuracy of how many times participants correctly
recognised a change in either name or occupation is the dependent variable.
SPSS 26 was used for the analysis of the data.
Participants:
We studied consenting participants comprised of 190 (27 males,160
females with three participants not stating their gender) postgraduate masters
psychology student. All enrolled in the University of Bed between 2018 and 2020
at the London campus. The participants' ages ranged between 19 and 57 years.
The participants were recruited via emails sent out to the entire class after
introducing the topic and experiment during a face to face lecture on campus at
the psychology department.

7
Materials:
Demographic characteristics
Relevant data was gathered through the use of using Qualtrics survey
software. The survey has two phases. The survey exposes and tests the
participants to pictures of 30 digital images of faces from the Karolinska Directed
Emotional Faces database (Lundqvist et al., 1998). The pictures consisted of all
males wearing grey t-shirts. The pictures had their eyes and mouths fixed to the
same coordinates for image standardisation. Also, the pictures were converted to
greyscale using IrfanView 4.37. the sizes of the pictures were fixed at 14 x 20
centimetres (cm). The names were put above the image and the occupation
below it. Every single face had a unique name and occupation.
Procedure:
The participants were recruited via posters, leaflets, and Email. The
students were then introduced to the study during a face-to-face lecture and
emails sent out to all the students. They were asked to access the survey online
with the link provided. They were informed of the procedure, and consents were
taken online before the survey could be accessed. Participants were advised that
they could withdraw at any time, and their data kept confidential. Incentives were
given, in the form of either amazon £10 gift card or a chance to win £100 from a
raffle draw, depending on their choice. All students participated (100% feedback).
The survey consisted of two timed phases. The first phase of the survey
was the exposure phase. The participants were given 30 male faces with

8
associated names and occupations on slides and asked to memorise. Each slide
was shown on the screen for 5000ms, and a blank white image was placed
between each slide for a duration of 1000ms.
Then they moved to the second phase. This phase assessed the recall
ability and recognition accuracy and saw any changes in the names and
occupations attached to the images shown. The 30 faces were shown again. Ten
of the images were unchanged from the first phase. Ten had different names,
and the remaining ten had a different occupation. The participants were given
instructions to indicate using the appropriate keyboard response, whether they
had seen the same information in the first phase or whether there was a change
in terms of the name and occupation.
After the responses were collated and analysed, a descriptive and
inferential analysis was completed, and the results were interpreted. The
participants were debriefed about the study. The analysis was done in line with
the hypothesis that a person's name is more difficult to remember than their
occupation.

9
Results
The study set out to investigate one hypothesis. A lower number of
participants would identify the name change, and a higher number of participants
would identify the occupation. The data was analysed by doing a descriptive and
inferential analysis (paired T-test).
The study had a 100% feedback with 190 responses. The participants
consisted of 27 male (14.2%), 160 female (84.2%) and three participants gender
were not recorded. The age range was between 19 and 57, with an average age
was 27.5 years and a standard deviation of 8.89. Seven participants ages were
missing.
A paired-samples t-test indicated that scores were significantly higher for
those that identified a change in occupation subscale with an average of 7.46
and a standard deviation of 2.47. However, the subscale of those who identified
a name change with an average of 4.47 and a standard deviation of 2.17. The
equation for the test is t(189) = -14.69, p < .001, d = 1.29. their mean difference
can be seen along with the level of confidence in the figure below.

10
Name Occupation
0
1
2
3
4
5
6
7
8
4
7
Mean
Figure 1 shows the mean difference between participants accuracy to identify changes in the
name or occupation.

11
Discussion
The results of the experiment in terms of consistency support the trend
suggested in the literature by researchers such as Bruce & Young (1986), Terry
(1994) and McCluney & Krauter (1997). Bruce & Young suggested that there was
a lower accuracy when trying to recall names than with occupation. Similar to the
suggestions made by Terry (1994) and McCluney & Krauter (1997).
The results of the analysis followed the trend previous found. However,
unlike previous studies done with familiar faces, this study was done with
unfamiliar faces. The result proves that for both familiar and unfamiliar faces, the
results are the same. There might not be as significant a difference in the
process of face recognition as previously suggested.
Although this study had reduced participation from males which might
have caused some degree of bias, some studies have shown that females have
a higher recall (Herlitz et al., 1997; Janssen & Murachver, 2014; Mishra et al.,
2019). Hence in future studies, robust recruitment must be done to verify if
gender is not a factor contributing to the higher averages seen in identifying
occupation.
Also, there were noticed to missing data which though would not have
made a significant effect. However, future research needs to consider the
possibility of these types of errors and avoid them. This study used a survey
approach that may not be the most appropriate for this type of experiment.
However, it is the most convenient method of collecting the data. A standardised
method of assessing this type of experiment needs to be developed.

12

13
References
Bruce, V., & Young, A. (1986). Understanding face recognition. British Journal of
Psychology, 77(3), 305–327. https://doi.org/10.1111/j.2044-
8295.1986.tb02199.x
Goldstein, H., & James, A. N. (1983). Efficient estimation for a multiple matrix
sample design. British Journal of Mathematical and Statistical Psychology,
36(2), 167–174. https://doi.org/10.1111/j.2044-8317.1983.tb01122.x
Herlitz, A., Nilsson, L. G., & Bäckman, L. (1997). Gender differences in episodic
memory. Memory and Cognition, 25(6), 801–811.
https://doi.org/10.3758/BF03211324
Janssen, A., & Murachver, T. (2014). the Role of Gender in New.
Lundqvist, D., Flykt, A., & Öhman, A. (1998). The Karolinska Directed Emotional
Faces - KDEF. CD ROM from Department of Clinical Neuroscience, P,
91(630), 2–2. https://doi.org/ISBN 91-630-7164-9
McCluney, M. M., & Krauter, E. E. (1997). Mr. Barber or a Barber: Remembering
Names and Occupations. Psychological Reports, 81(3), 847–863.
https://doi.org/10.2466/pr0.1997.81.3.847
Mishra, M. V., Likitlersuang, J., B Wilmer, J., Cohan, S., Germine, L., & DeGutis,
J. M. (2019). Gender Differences in Familiar Face Recognition and the
Influence of Sociocultural Gender Inequality. Scientific Reports, 9(1), 1–12.
https://doi.org/10.1038/s41598-019-54074-5
Severin, F. T., Eysenck, M. W., & Keane, M. T. (2005). Cognitive psychology: A
student's handbook, 5th ed. In Cognitive psychology: A student's handbook,

14
5th ed. (Vol. 17, Issue 4). Psychology Press.
https://doi.org/10.5840/schoolman194017410
Terry, R. L. (1994). Effects of facial transformations on accuracy of recognition.
Journal of Social Psychology, 134(4), 483–492.
https://doi.org/10.1080/00224545.1994.9712199