Second Article: Suspicion and Eyewitness IDs

Neuschatz, J., Lawson, D., Fairless, A., Powers, R., Neuschatz, J., Goodsell, C., & Toglia, M. (2007). The Mitigating Effects of Suspicion on Post-Identification Feedback and on Retrospective Eyewitness Memory Law and Human Behavior, 31 (3), 231-247 DOI: 10.1007/s10979-006-9047-7

Okay, now that we're getting into eyewitnessing I decided to post this one about identification instructions. For all you future lawyers, this has much more practical applications than last week's. Plus, since I know you're all studying furiously for the midterm, it's shorter, too. It's illustrative of a lot of the research done on eyewitness procedures in psychology, however.

One more nota bene before I launch into it: most of the authors are working out of the University of Alabama, which was one of the first American universities to offer a doctoral program for clinical psychologists working in the criminal justice system. It was set up by Stanley Brodsky, former head of the American Association of Correctional Psychologists and the man who turned Correctional Psychology into a fully-defined and accepted professional field in the 1970's.

Returning to the present, this study addresses one of the major problems encountered in eyewitness identification procedures carried out by the police: feedback. In 1999 a group of eminent psychologists, lawyers, and police officials were assembled by the Justice Department in order to come up with a guide for police on how to handle eyewitness testimony to ensure its reliability and admissibility. This resulted in "Eyewitness Evidence: A Guide for Law Enforcement." It covered most of the elements that affect the veracity of eyewitness testimony and laid out interviewing tactics that should be used. But, as Prof. Geiselman referenced, this is one of those books that gets tossed into a drawer in most precincts without being read (though there are several cities, such as Minneapolis, where these guidelines have been taken to heart and implemented in the force). One of the major sticking points in creating the Guide, though, was feedback. The psychologists, being scientists, insisted on a double-blind procedure where the officer administering the lineup was unfamiliar with the case and couldn't give any feedback, consciously or unconsciously, to the eyewitness. This was hard for the police to accept, though, since they took it as an accusation of dishonesty or attempting to sway witnesses. This dispute wasn't resolved, and so nothing about blind lineups was put in the Guide, so this is still very much an issue that anyone who deals with eyewitnesses has to confront.

The current study looks at two different methods for reducing feedback effects: implanting suspicion and the interestingly-titled "confidence prophylactic." Suspicion was used by telling the participants that the administrator was telling everyone they were right, regardless of their choice. The confidence prophylactic has participants make ratings about the circumstances of their initial viewing of the suspect (e.g. how good the lighting was, their distance, etc.) and their confidence in their identification. The idea behind this is that making eyewitnesses make judgments about how good their memory for the incident really is should help them to calibrate their confidence more accurately, and prevent the inflation seen after confirming feedback.

The second manipulation in this study is the time between the identification and the intervention. Experiments 1 and 2 looked at suspicion, with an immediate delay and experiment 1 and a week-long delay in 2. Experiment 3 examined the confidence prophylactic with and without a delay. Ultimately, they found that suspicion is effective in reducing confidence inflation regardless of when it's given, but that the prophylactic, living up to its name, is only effective if it's used at the time of the ID.

This paper is pretty representative of a lot of the eyewitness work done in psychology, and it raises some possible issues. For instance, most studies focus on reducing false positives: identifying a suspect who isn't the perpetrator. For this reason they use target-absent lineups, where any identification made is sure to be wrong. Are there any advantages to this procedure over a target-present lineup? And is this approach to research better than focusing on minimizing false negatives: failing to identify a perpetrator? Furthermore, how practical is the suspicion intervention? It might not be practicable, and certainly wouldn't be well-received in the precinct, to tell every witness that the police officer who just showed her the suspects might have been lying to her.