top of page

Thanks to everyone who took part in this project. In this task, you viewed 60 pairs of images and were asked to decide whether you thought they were of the same person or not. If you have your e-mail with your scores on this project on hand you should be able to compare them with the full list of participants who competed this research at the bottom of this report. Some of you reported being given scores of more than 60 out of 60 on this test. This was an error and we have added your correct scores in a list below in Table 3. Sorry about the confusion this may have caused. The main aim of this project was to pilot test a brand new Chinese Face Matching Test for a Singapore project, and to see how scores compared with those on the Cambridge Face Memory Test: Extended (CFMT+) (Russell, Duchaine, & Nakayama, 2009) and the Glasgow Face Matching Test (GFMT) (Burton, White, & McNeill, 2010). This gives us better insight into the other-race effect in face recognition, which suggests that “identifying a person from a different ethnic group results in even poorer performance [compared to identifying a person within one’s own ethnic group]” (Robertson Black, Chamberlain, Megreya, & Davis, 2020, p. 206). A summary of this research can be found in this article by David Robertson, Ahmed Megreya, and Josh Davis in the Conversation (Robertson, Megreya, & Davis, 2020).

In total, 251 volunteers completed the test (aged 20-76 years, M = 38.4, SD = 9.98, male = 64, female = 184, White = 139, Chinese = 67, other = 45). We were able to match all but three of those participants with their original scores on the CFMT+ and the GFMT stored on our database. Therefore, data from 248 participants are reported here. Table 1 displays median and mean scores on each test.

To see how you scored against the other participants, Figure 1 displays the frequency of total scores for the Chinese Face Matching Test. Table 2 displays the Spearman’s correlation coefficients between total scores on each test, while scatter diagrams in Figure 2 and 3 visually display the relationships between each test.

As indicated by the correlations, people who achieve high scores on the GFMT, also do very well on the CFMT+. There is, however, a stronger correlation between the GFMT and Chinese Face Matching Test, not surprising perhaps, as both of these tests measure simultaneous face matching, than correlations between these matching tests and facial memory as measured by the CFMT+.

Despite these expected results, as can be seen, scores on the Chinese Face Matching Test were extremely high and as the majority of participants were not of Chinese ethnicity, this suggests the test was probably too easy and suffered from ceiling effects. We will therefore be developing a new version of this test with the inclusion of harder trials.

Figure 1 Frequency of scores on the Chinese Face Matching Test

Figure 2 Scatterplot diagrams showing the relationship between the Chinese Face Matching Test and the Cambridge Face Memory Test

Figure 3 Scatterplot diagrams showing the relationship between the Chinese Face Matching Test and the Glasgow Face Matching Test

The scores – in participant code order can be found below for the Chinese Face Matching Test. Note – these are the codes entered into the system. Three, however did not match with our database.

Table 3 Scores for the Chinese Face Matching Test


Burton, A. M., White, D., & McNeill, A. (2010). The Glasgow face matching test. Behavior Research Methods, 42(1), 286-291. DOI: 10.3758/BRM.42.1.286

Robertson, D. J., Black, J., Chamberlain, B., Megreya, A. M., & Davis, J. P. (2020). Super‐recognisers show an advantage for other race face identification. Applied Cognitive Psychology, 34(1), 205-216. (pre-print here). DOI: 10.1002/acp.3608

Robertson, D. J., Megreya, A., & Davis, J.P. (2020, January 10). Facial recognition: research reveals new abilities of ‘super-recognisers’. The Conversation. Retrieved from

Russell, R., Duchaine, B., & Nakayama, K. (2009). Super-recognizers: People with extraordinary face recognition ability. Psychonomic Bulletin & Review, 16(2), 252-257. DOI: 10.3758/PBR.16.2.252


Recent Posts

See All


bottom of page