‘The witness is lying!’: the impact of a defendant countering a jailhouse informant’s testimony

ABSTRACT

The present study investigated the impact of a defendant explicitly countering the claim of a jailhouse informant that the defendant confessed to a murder. Jury-eligible community members (N = 127) recruited from Amazon’s Mechanical Turk read (via Qualtrics) a fictional trial summary in which a defendant was accused of murder. The trial was presented in one of four conditions: jailhouse informant only, defendant explicitly countered the jailhouse informant’s testimony (said he did not talk to the jailhouse informant or they talked but not about the case), defendant testified but did not explicitly counter the jailhouse informant’s testimony (simply said he was innocent), and no jailhouse informant testimony. As predicted, when the defendant explicitly countered the jailhouse informant’s testimony the number of guilty verdicts was lower than that when the jailhouse informant only testified. In addition, there was significant mediation, such that explicitly countering the jailhouse informant led to lower jailhouse informant credibility and fewer guilty verdicts. Finally, cognitive network analyses showed that countering the jailhouse informant’s testimony was critical to participants representing the jailhouse informant’s testimony in a negative light. We discuss these results in terms of the importance of countering what is often unreliable jailhouse informant testimony.

KEYWORDS:

• Jailhouse informant
• defendant
• murder
• law

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 Note that we also conducted the same analyses with 30 (24%) participants excluded for not correctly answering one of four manipulation-check questions (did jailhouse informant/defendant testify, age of victims, and type of trial condition). Across 30 regression analyses, the excluded and total samples had the same significant results, except for four comparisons (87% of the analyses had the same results). Three of these four involved the exploratory no-mention condition. In addition, three of the four comparisons were significant for the total sample but not the excluded sample—all non-significant comparisons were in the same direction as the corresponding significant result. Thus, we chose to present the total sample because we felt the full sample more accurately reflected how jurors in the real world may forget certain aspects of a case but still render a decision.

2 Note that the degree of freedom for the denominator are less for the defendant dependent variables compared to the jailhouse informant dependent variables. Besides differences in the number of participants per type of trial condition, this difference in degrees of freedom is the result of 21 participants not correctly recalling that the defendant testified, and thus not answering rating questions about the defendant.