To what extent is the cognitive process of memory reliable?

A schema is a mental representation about the world and in order to reconstruct memory you need to activate the schema relevant to that event. Researchers have investigated whether this schema activation and memory reconstruction is reliable, especially when used in eye-witness testimonies. Some studies have suggested that the cognitive process of memory is not reliable. For instance, Loftus and Palmer (1974) examined the influence of leading questions on the testimonies of eye-witnesses. They showed participants a video clip of two cars crashing and then asked them to estimate the speed the cars were going. However, they changed the verb of the question for each group; for instance, one group was asked "How fast were the cars going when they bumped into each other?", while another was asked "How fast were the cars going when they slammed into each other?". Their hypothesis was that the "slammed" leading verb would activate a relevant schema in the participant's minds, and they would overestimate the speed of the cars. This was actually what they found, suggesting that eye-witness testimonies are unreliable and can be influenced by leading questions. Nevertheless, while the results of Loftus and Palmer's (1974) study suggest that memory is not reliable, there are some limitations to be considered. Firstly, the controlled conditions of the experiment reduce the ecological validity (applicability to the real world), while the validity of the results is also decreased by the fact that there was no emotion present, which is often experienced by real life eye-witnesses. One study that attempted to address these limitations was conducted by Yuille and Cutshall, who attempted to discern whether leading questions would affect memory of eyewitnesses at a real crime scene. The researchers contacted people who had been eyewitnesses to a real shop theft and shooting after 4 months and asked them a series of leading questions that were not true; "was there a broken headlight on the car?", "was there a yellow panel on the car?". Surprisingly, the participants' answers were a lot more reliable than expected. This suggests that in real life settings, memory might be more reliable than thought before. Nevertheless, this is a hard study to replicate because it was a field study (done in real world settings) and there are many possible explanations as to why the participants remembered information more accurately. For instance, it could be due to them forming a flashbulb memory, which occurs when a strong emotion is associated to a memory and thus makes it a lot more easily remembered. These types of memories are different to normal memories, so more research is needed to determine whether memory is truly reliable in real world settings.