The model suggests that memories are built up through the networking of links and associations between nodes. Inconsistent information results from wrongly linked nodes. This information will receive more attention therefore its linkages will be strengthened resulting in improved recall of the said information.
The second model, the Schema theory, takes a different approach. It states that we use our past experiences to help with the comprehension of the world. Past experiences form cognitive structures which are then used as a basis for what we experience and what we plan in the future. Incorrect memories arise form personal reconstruction to make sense of things which may not be true.
To demonstrate this concept I will give an example of the real world functioning of schemas. If a person walking own the street sees a person trying to force a side door to a house it is possible that their schema will tell them that this person is a criminal. This is because cognitive structure based on what the person may have hear or previously experienced says that this is what criminals sometimes do. What the man is really trying to do is to force to the door because it is jammed.
The Association theory takes the bottom up approach, namely that our memories are constructed form the networking of information whereas the Schema theory is a top down model in that it relies on our past experiences to formulate schema that influence our perception and memories.
Much of our current understanding into the accuracy of eye witness accounts comes from the studies conducted by Elizabeth Loftus in the 1970’s. Loftus investigated the effects on leading questions and the vocabulary used in formulating questions. It was found that leading questions i.e. questions that identify certain objects can dramatically alter the memories of a witness. If false information is contained within the question then this information can become incorporated into the memory. “What was he wearing on his right wrist” is an example of a leading question which could then cause the witness to invent something when in reality there was nothing on his right wrist.
The ‘strength’ of vocabulary can also affect the response as was found found by Loftus and Palmer (1974) when they carried out a study which required the participants to watch a video of a car crash and then respond to various questions. A key question “what speed were the cars travelling when they…” was asked and the blank was filled with words of varying intensity, e.g. “bumped”, “smashed”, “collided”. It was found that the greater the word intensity, the higher the speed.
Whilst it was clear that these studies point out inaccuracies in eyewitness testimony they deal only with peripheral details about the event. It was found that the central details remained unchanged but that the ability to take on new information which could modify the existing memory increased over time, so that the further away the event was the more liable the witness to recall misinformation.
These studies deliver an insight into the inaccuracies of eye witness testimony however one could argue that they lack any real world application as the laboratory environment is vastly different from the real world and the speed at which the speed and method of information delivery is far slower than in real life. Also in a real life car crash the witness would not be wholly concentrating on the crash when it happens. In laboratory studies the same item is used for stimulus and recall whereas in reality the recall item could be a photo where the situation is different e.g. wearing different clothes.
