Identifying individuals from one’s own ‘group’ is easier than identifying someone from outside the ‘group.’  If you were to stop me and say “but that’s obvious, it’s always easier to identify people you already know than identifying a stranger” I’d have to respond that “you mis-read what I wrote.  ‘Group’ does not mean ‘friends’ or ‘acquaintances;’ it means a social category such as those of the same race or ethnicity.”

Don’t take my word for it – this is well-established social science.  In BLINDSPOT, researchers Mahzarin Banaji and Anthony Greenwald write of how

[t]his ability to perceptually discriminate between two or more members of a familiar category extends to the social category of race:  By nine months of age infants are better able to tell two one-race faces apart compared to two other-race faces.

The authors explain that this is known as the “out-group homogeneity effect.”

And you need not restrict yourself to social scientists’ findings; this has now been set as a principle of law in New York State.  In late December, 2017, the New York State Court of Appeals wrote the following:

In light of the near consensus among cognitive and social psychologists that people have significantly greater difficulty in accurately identifying members of a different race than in accurately identifying members of their own race, the risk of wrongful convictions involving cross-racial identifications demands a new approach. We hold that when identification is an issue in a criminal case and the identifying witness and defendant appear to be of different races, upon request, a party is entitled to a charge on cross-racial identification.

People v. Boone, 2017 NY Slip Op 08713, ¶¶ 1-2.

Of concern to the New York Court, beyond the risk of error in such cases, was the absence of knowledge of this problem among jurors.  “[W]hile the cross-race effect is a matter of common sense and experience for some jurors, it is by no means a universal belief shared by all. The need for a charge on the cross-race effect is evident.”

The bottom line for the New York Court was that increasing reliability in jury determinations and avoiding wrongful convictions warranted judicial intervention.  The Court concluded that regardless of whether an expert testified or whether the defense aggressively cross-examined the ‘certain’ eyewitness, the cautionary jury instruction was required, explaining that

the jury should consider whether there is a difference in race between the defendant and the witness who identified the defendant, and (2) that, if so, the jury should consider (a) that some people have greater difficulty in accurately identifying members of a different race than in accurately identifying members of their own race and (b) whether the difference in race affected the accuracy of the witness’s identification.

People v Boone, 2017 N.Y. LEXIS 3722, *17-18, 2017 NY Slip Op 08713, 7-8.

The bottom line?  This is a Court that takes seriously the notion of fair play, and that when jurors don’t know basic science their weighing of evidence is impaired and the resulting verdict is less valid.