Influencing what an eyewitness remembers through gesture

We know that the way we word a question can influence the answer we get. If I ask, for instance, "how tall was the man?" you may respond differently than if I asked "how short was the man?" This is because the wording of the second question encourages you to think of the man as being shorter than average. This is an important thing to consider when questioning eyewitnesses (who we know can be very sensitive to leading questions). A police interviewer cannot say "did the man have a beard?" as the mere suggestion of a beard may bias the witness into thinking there was one. Because of this, police interviewers need to be very careful that they don't implant suggestions into eyewitnesses when questioning them.

However, while we can be very careful about what we say in our speech, we have little control over what we 'say' with our hand gestures. We know that people can pick up on important information in gestures (even if the listener isn't aware they have) - so can hand gestures influence eyewitnesses too?

This is exactly what I set out to test in one of my first studies. I got 66 participants to watch a piece of crime scene footage, and then questioned them on what they could remember afterwards. During questioning, the 'witnesses' either saw the interviewer performing a gesture that conveyed something true in the video, or something false. For instance, in the video the man was wearing a ring. So, people either saw the interviewer gesturing a ring when asking "did you notice any jewellery?" (the true gesture) or they saw him gesturing a watch (the false gesture). (I also used a control group - no gestures - to see what people would say without any gestures influencing them).

The 'witnesses' either saw the true 'ring' gesture (left) or the false 'watch' gesture (right).

So, did the gestures influence what the 'witnesses' remembered? Well, 95% of people that saw the ring gesture got the answer right (they said 'ring) compared to those who saw the 'watch' gesture (where only 67% said 'ring') and the control group. So, the true 'ring' gesture did help people toward the right answer.

Also, the 'watch' gesture misled people: People that saw the 'watch' gesture were more likely to say that they remembered the man wearing a watch (which made up 75% of incorrect answers). What's even more surprising is that some people could even remember what colour the watch was!

Obviously, this is very important - even if a police interviewer controls what they say and asks a question that shouldn't be leading in any way, their hand gesture can still 'give away' an answer, possibly without them even knowing. While police interviews are audio recorded, without any video footage we have no idea what gestures have been produced and whether this may have influenced the eyewitness!

References:
Gurney, D. J., & Pine, K. J. (2011). The gestural misinformation effect. Under review.
Gurney, D.J., & Pine, K. J. (2009). Can misleading hand gestures influence? In Cognitive Psychology Section 26th Annual Conference, 1-2 September 2009 (p.44) University of Hertfordshire, Hertfordshire, UK: British Psychological Society.

Comments