Non-Stranger Identification: How Accurately Do Eyewitnesses Determine If a Person is Familiar

Abstract:     
The extensive research literature on eyewitness memory has been restricted to memory for strangers. This study assessed how accurately people can recognize individuals they have casually seen before, and whether this determination is affected by the cross-race effect and exposure time to the test picture. One hundred and fifty seven Asian and non-Hispanic White high school students in a small private high school viewed yearbook pictures of (a) graduated students who were seniors when participants were freshman (“familiar”) or (b) unfamiliar individuals, and responded whether each target individual was familiar. Recognition accuracy was surprisingly low. The mean hit rate was .45. The mean false alarm rate was .28; 28% of unfamiliar individuals were misidentified as familiar. Further, the cross-race effect was not evidenced in these results. An eyewitness’s report that he can recognize a perpetrator because he has seen him casually in the past is of dubious reliability.

Number of Pages in PDF File: 23

Keywords: eyewitness memory, criminal law

JEL Classification: K14

working papers series