認知心理学6. Eyewitness testimony: 目撃証言
Today
- Where we’ve been so far
- Considering what we know about human information processing, how reliable is eyewitness information actually?
The Innocence Project
- 212 exonerations since 1992
- 75% of those wrongful convictions due to the misidentification from ''eyewitnesses''.
- How misidentifications can happen:
- at encoding
- at maintenance
- at retrieval
Potential pitfalls of misidentification
| Name | Charges | Yr. Convicted | Sentence/ Time Served | Evidence for Conviction |
|---|---|---|---|---|
| K. Adam | 2 counts murder, Rape | 1978 | 75/16 | Witness id |
| G. Alejandro | Sexual Assault | 1990 | 4- Dec | Blood evidence, Victim id. |
| K. Bloodsworth | Murder, Rape | 1985 | Death, reduced to life/ 9 | 5 witness id’s, statements |
| R. Bullock | Aggravate sexual assault | 1984 | 60/10.5 | 2 victim id’s. Police id |
| D. Brison | Rape & Kidnapping | 1991 | 42/ 3.5 | Victim Id, Hair analysis, Weak alibi |
How misidentifications can happen
- At encoding
- Attention limitations
- change blindness
- Improper feature binding
- 知ってる人ほど記憶と結びつきやすい。やったのはあいつだ。って思う。
At the level of encoding misidentification is due to attention limitations.
I have clip here which should hopefully demonstrate how attentional resources at encoding level influences what we are aware of.
So this video was an example of change blindness.
Change Blindness occurs when a person is viewing a scene but fails to notice a change in the scene. For some of you, focusing your attention on counting how many passes the players made caused you to miss the gorilla walking through the middle.
Another thing that can happen during encoding is improper feature binding. Im rpo
How misidentifications can happen
- At maintenance/ retrieval
- Misinformation effect (Loftus, 1978)
- Video of car running stop sign and causing an accident
- Accurate description or description that mentioned a yield sign.
- People hearing the altered version said they saw a yield sign more often than the others.
Loftus, Miller, & Burns（1978）は自動車事
故の目撃場面を模した実験を行い，目撃記憶の正確さを検討した。
彼女らによる実験では，事故場面の交通標識を被験者がどれだけ正確に覚えているのかが検討された。
一連の画像が被験者に提示された後，画像の内容に関する質問用紙を被験者に配り，回答を求めた。
一部の被験者には，画像の内容と一致しない誤情報を含んだ質問文が示された。
たとえば，“前方優先道路”を示す道路標識が現れた交差点場面についての質問文に，画像の内容とは異なる情報である“一時停止の標識”が記述されていた。
引き続き実施された再認記憶テストでは，2枚1組の画像が提示され，どちらが画像提示場面で見た画像であるかを判断することが被験者に求められた。
被験者に提示された一連の画像対の中には，最初の画像提示で現れた画像と，質問文中の誤情報と一致する画像とを組み合わせた1対が含まれていた。
たとえば，前方優先道路の標識を含む画像（最初の提示された画像）と，一時停止の標識を含む画像（質問文の誤情報と一致）が組み合わされていた。
誤情報を含んだ質問文を提示された被験者が誤情報を含む画像を選択した割合は .75になり，誤情報を提
示されなかった被験者が誤情報を含む画像を選択した割合は .41であった。
質問文中に含まれていた誤情報の影響を受けた被験者は，実際には見ていない道路標識を見たと報告することが明らかになった。
(再認記憶とソースモニタリングにおける誤情報効果)
At the retention and retrieval levels of processing misidentifications can happen due to misinformation. This effect occurs when misinformation influences your memory for an event.
In Loftus’s study subjects watched a video of a car running a stop sign and causing an accident.
After watching the video they were given either an accurate description of the accident or a altered version which replaced the stop sign with a yield sign. Thos subjects who heard the altered version reported seeing the yield sign in the video more often than others.
This is a good example as to how false information of an already experienced event can alter your memory for that event.
How misidentifications can happen
- Brown, Deffenbacher, & Sturgill (1977)
- A witness in a rape case was shown a photo array where only one photo - of the person police suspected was the perpetrator - was marked with an “R.”
- Witnesses substantially changed their description of a perpetrator (including key information such as height, weight and presence of facial hair) after they learned more about a particular suspect.
- Witnesses only made an identification after multiple photo arrays or lineups - and then made hesitant identifications (saying they “thought” the person “might be” the perpetrator, for example), but at trial the jury was told the witnesses did not waver in identifying the suspect.
フォトラインナップでマークがついてる物に無意識のうちに目撃情報を変えていく。
最初は自信なくても、何度もやっているうちにすごく自信になる。
Here is a real example of how misidentifications occur at the retention and retrieval levels.
Relative Judgments
- relative judgment
- absolute judgment
- Wells (1993)
- Ss were witnesses to a staged crime.
- 1/2 the Ss were shown a line-up with suspect present.
- 1/2 the Ss were shown a line-up with the suspect not present
- Ss were witnesses to a staged crime.
| 1 | 2 | 3(本物) | 4 | 5 | 6 | No choice | |
|---|---|---|---|---|---|---|---|
| *Suspect present | 3% | 13% | 54% | 3% | 3% | 3% | 21% |
| *Suspect Absent | 6% | 38% | ---- | 12% | 7% | 5% | 32% |
They were all told that the suspect may or not be present in the line-up
If eyewitness identification is a result of true recognition then we would expect that the Ss who were shown the line-up with the suspect present would have a higher rate of identification then those who saw the line-up without the suspect present. And those who say the line-up without the suspect present should respond none of the above to a greater extant
One reason subjects are making relative judgments is when witnesses are shown several photos at once their attention is focused on comparing the photos of the line-up with each other as well as their memory. Attention is limited when we extend the limits of our resources errors increase.
Sequential Line-ups
- Lindsey & Wells (1985)
- Ss were shown a photo line-up in a sequential fashion
- Results:
- Ss more sensitive to the present vs. absent instructions.
- Target present: Correct identification increased and misidentification decreased.
- Target absent: Misidentification decreased
One way to reduce false positives or misidentification is to present the photos one at a time.
Before moving onto the next photo they had to decide whether or not the person they had just seen was the perpetrator. Since they have not seen the other photos yet they were unable to make a relative judgment.
Thus, subjects had to compare each suspect with their memory and were unable to make comparisons across suspects. Although the could compare the current photo with a previous photo, because they had already made a judgment it didn’t interfere with the comparison process.
Guidelines
- person conducting the line-up should be unaware as to who the suspect is
- Witness is to be told that the suspect may or may not be present
- Suspect should not "stand out" in the line-up
- Witness should make a clear statement before receiving any feedback.
Improving Eyewitness Recall：Geiselaman (1985)
- Geiselman et. Al. (1985).
- subjects view a 4 min video of a crime
- Standard / hypnosis / cognitive
- Cognitive Interview: Four memory retrieval techniques.
- Reinstate the context of the incident
- Report everything
- recall of event in different order
- Recall the event from different perspective.
Geiseiman(1985)は検索に関する知見に基づいて4つの原理からなる面接法を考案し、それを認知面接（cognitive interview）とよんだ。4つの原理とは次のような物である。
- 事件が目撃されたときに、その周囲の環境並びに目撃者当人の文脈を心理的に復元してみる。目撃者をさかのぼって考えさせ、直前の起こったこと、目撃者自身の行動、気分を再生させる。
- どんな些細なことでもかまわないから、全てのことを報告するようにさせる。
- 時間的に順を追ったりさかのぼったり、いろいろな順序で事件を記述するように求める。
- 異なった視点から事件を述べてみるように求める（例：道路の反対側に立っていたとしたら、何を見たと思いますか）
これらの原理は可能性のある検索ルートの数を最大にするように考えられている。
基本的な考えは、文脈の再活性化は元の事件の記憶の手がかりになるという物である。
ガイゼルマンの方法だと、標準的な方法に比べて３０％以上も正しく情報が再生された。
(日常記憶研究の動向)
The value of hypnosis within the legal system is debatable. There are concerns over false memories being implanted with the witness and concerns about the person performing the hypnosis biasing the memory of the witness. Furthermore, because people undergo hypnosis more than once its possible that it is the repeated testing rather than the hypnosis itself that is responsible for additional recall.
However, hypnosis has been successfully used in court to elicit more detail from witnesses about a particular event. And research has show it to be fairly reliable.
In a study by Geiselman an colleagues subjects viewed a 4min video of a violent crime, and were then interviewed a few days later by police.
There were three interview conditions. The standard interview in which the subjects were questioned in the standard way police do.
The hypnosis condition in which hypnotized subjects were asked to restate what they saw in the film.
And the cognitive interview.
- Results:
- Both hypnosis and cognitive better recall
- No difference in recall b/w hypnosis and cognitive interview.
The results demonstrated that both the hypnosis and cognitive interviews produced better recall than the standard interview. But there was no difference in recall between the hypnosis interview and the cognitive interview.
What this tells us is that hypnosis is a reliable way to elicit additional recall. However, cognitive interviews work just as well and is easier to learn and administer than hypnosis. Additionally, cognitive interviews don’t raise the concerns hypnosis does, and thus juries are likely to be less skeptical of events that were recalled using cognitive techniques.
