Model Essay
To what extent is one cognitive process reliable? Support your answer with reference to relevant psychological research.
One cognitive process whose reliability has been widely debated is memory. In psychology, reliability refers to the extent to which a cognitive process produces accurate and consistent representations of events. Memory is often treated as unreliable because it is reconstructive rather than reproductive: people do not retrieve an exact copy of the past, but rebuild it using schemas, expectations and post-event information. However, the evidence does not support the claim that memory is simply inaccurate. A balanced judgement is that memory is reliable only to a moderate extent: it is vulnerable to distortion under some conditions, especially when misleading information is introduced, but it can be highly accurate when the event is meaningful, emotionally salient and recalled in ecologically valid contexts.
Bartlett’s theory of reconstructive memory provides a useful explanation for why memory may be unreliable. Bartlett argued that recall is shaped by schemas, which are organised packets of knowledge based on prior experience. Because new information is interpreted through existing schemas, memory can become distorted over time. His classic study, War of the Ghosts, demonstrated this. British participants read an unfamiliar Native American folk tale and were later asked to reproduce it. Their recollections became shorter and altered in ways that made the story more conventional and understandable within their own culture. Unfamiliar details were changed or omitted. This suggests that memory is not a faithful recording of events, but is reconstructed in ways that fit existing knowledge. In terms of reliability, Bartlett’s work shows that recall is susceptible to systematic distortion rather than random forgetting.
This theoretical argument is strongly supported by Loftus and Palmer’s research on leading questions. Their aim was to investigate whether post-event information, specifically the wording of a question, could affect eyewitness memory. In one experiment, participants watched films of car accidents and were then asked how fast the cars were going when they “hit”, “bumped”, “collided”, “smashed” or “contacted” each other. Estimated speeds were higher when the verb was more forceful, with “smashed” producing the highest estimates. In a second experiment, participants who had previously heard the word “smashed” were more likely to report seeing broken glass, even though there was none. This is powerful evidence that memory is unreliable because recall can be reshaped by leading questions after the event itself.
The significance of Loftus and Palmer is not just that memory can be inaccurate, but that the inaccuracy is socially produced. Post-event information can become integrated into the memory trace, making eyewitness testimony potentially unreliable. This has clear real-world implications for police interviews and courtroom evidence. The study also supports the reconstructive view of memory: participants did not simply report what they had seen, but reconstructed the event using cues provided by the researcher.
At the same time, the extent to which this proves memory is unreliable should be questioned. Loftus and Palmer was a laboratory experiment using film clips of accidents rather than a genuinely stressful experience. This increases internal validity because the wording of the question was tightly controlled, but it reduces ecological validity. Watching a short video in an experiment is not the same as witnessing a serious crime. Participants may also have been affected by demand characteristics, guessing that the wording of the question was important. Therefore, while the study shows that memory can be distorted, it may exaggerate unreliability compared with real-life memory.
This is where Yuille and Cutshall provide an important counter-argument. They studied eyewitnesses to a real shooting in Vancouver and interviewed them several months later. The witnesses were found to be highly accurate in their recall of the incident, and misleading questions did not substantially distort their memories. This suggests that memory can be reliable, especially when the event is significant and emotionally arousing. Unlike Loftus and Palmer, this was a real-world event with genuine consequences for the witnesses, so the findings have greater ecological validity.
Yuille and Cutshall therefore qualify the claim that memory is unreliable. It may be that memory for trivial, laboratory-based stimuli is more easily altered, whereas memory for highly salient events is retained more accurately. This can be linked to the role of emotion. High emotional arousal may strengthen memory consolidation, making recall more resistant to post-event misinformation. However, this conclusion should also be treated cautiously. Because Yuille and Cutshall studied a naturally occurring event, many variables could not be controlled. The sample was also relatively small and specific, making generalisation difficult. In addition, not all emotional memories are accurate; sometimes stress can narrow attention, leading to phenomena such as weapon focus. Even so, the study is valuable because it shows that the reliability of memory depends on context rather than being uniformly poor.
Further support for the idea that memory is sometimes reliable comes from the distinction between central and peripheral details. In emotionally significant situations, individuals may remember the central aspects of an event very well, even if peripheral details are less accurate. This means that asking whether memory is reliable requires nuance. Reliability is not absolute. It varies according to the type of memory being tested, the delay before recall, the presence of misleading information, and the personal meaning of the event.
Overall, memory is reliable only to a limited or moderate extent. Research such as Bartlett and Loftus and Palmer clearly shows that memory is reconstructive and vulnerable to distortion by schemas and leading questions. These findings make it difficult to regard memory as a wholly accurate cognitive process. However, evidence from Yuille and Cutshall suggests that under ecologically valid, high-stakes conditions memory can be surprisingly accurate and resistant to misinformation. Therefore, the best answer is not that memory is reliable or unreliable in general, but that its reliability is conditional. Memory is least reliable when recall is reconstructed from weak or ambiguous traces and most reliable when the event is meaningful, distinctive and retrieved without contaminating post-event information.
Tip:
A balanced judgement is that memory is reliable only to a moderate extent: it is vulnerable to distortion under some conditions, especially when misleading information is introduced, but it can be highly accurate when the event is meaningful, emotionally salient and recalled in ecologically valid contexts.
This gives you a clear, nuanced line of argument from the start. It directly answers the 'to what extent' demand and sets up the conditions under which you will judge memory as more or less reliable.
Bartlett’s theory of reconstructive memory provides a useful explanation for why memory may be unreliable.
This topic sentence does more than introduce Bartlett: it clearly ties the theory back to the question of reliability, so the paragraph stays argument-led rather than becoming a list of facts.
British participants read an unfamiliar Native American folk tale and were later asked to reproduce it.
This is a precise study detail that grounds your discussion in concrete research rather than vague reference. Including procedure like this helps the reader see exactly how the findings were generated.
This suggests that memory is not a faithful recording of events, but is reconstructed in ways that fit existing knowledge.
Here you move from description to interpretation by explaining what the distortions mean. The close reading of the findings is what turns the study into support for your argument about reliability.
This theoretical argument is strongly supported by Loftus and Palmer’s research on leading questions.
This is an effective bridge between paragraphs because it shows development in the argument: you move from theory to empirical support in a single, controlled sentence.
In one experiment, participants watched films of car accidents and were then asked how fast the cars were going when they “hit”, “bumped”, “collided”, “smashed” or “contacted” each other.
You select a highly relevant procedural detail and include the key manipulated verbs, which makes the study evidence specific and memorable. That precision strengthens the link to leading questions and post-event information.
This is powerful evidence that memory is unreliable because recall can be reshaped by leading questions after the event itself.
This sentence explicitly explains why the findings matter for the question. You are not leaving the study to speak for itself; you interpret its significance in terms of unreliability.
Their aim was to investigate whether post-event information, specifically the wording of a question, could affect eyewitness memory.
This is a strong argument move because you frame the study through its aim before giving details. That keeps the reader focused on the exact variable being tested and why it matters.
The significance of Loftus and Palmer is not just that memory can be inaccurate, but that the inaccuracy is socially produced.
This is a thoughtful evaluative shift: you deepen the argument by identifying the source of the distortion, not just stating that distortion happened. That conceptual step shows real control of the implications of the research.
Post-event information can become integrated into the memory trace, making eyewitness testimony potentially unreliable.
This precise psychological phrasing shows secure use of subject terminology. The sentence also neatly links cognitive theory to a practical consequence in eyewitness testimony.
This has clear real-world implications for police interviews and courtroom evidence.
You widen the discussion beyond the lab and show why the point matters in applied contexts. That helps the argument feel purposeful rather than purely abstract.
At the same time, the extent to which this proves memory is unreliable should be questioned.
This transition is excellent signposting because it signals a counter-argument and reminds the reader that the essay is weighing extent, not making a one-sided claim.
Loftus and Palmer was a laboratory experiment using film clips of accidents rather than a genuinely stressful experience.
This is a sharply chosen methodological detail because it sets up your later evaluation of how far the findings can be applied to real eyewitness memory.
This increases internal validity because the wording of the question was tightly controlled, but it reduces ecological validity.
You balance a strength against a limitation in one sentence, which is exactly the kind of measured evaluation strong psychology writing needs. The comparison shows you understand methodological trade-offs rather than treating validity as a simple good/bad issue.
Therefore, while the study shows that memory can be distorted, it may exaggerate unreliability compared with real-life memory.
This is an excellent qualifying judgement. You keep the value of the study while carefully limiting the scope of its conclusion, which is a sophisticated way to handle 'to what extent'.
This is where Yuille and Cutshall provide an important counter-argument.
This phrase clearly marks the essay’s shift to contrasting evidence. It helps the reader see that your discussion is deliberately balanced and cumulative.
They studied eyewitnesses to a real shooting in Vancouver and interviewed them several months later.
This concise detail establishes the study’s real-world context straight away. It gives the reader the essential procedural information needed to understand why this evidence contrasts with laboratory findings.
This suggests that memory can be reliable, especially when the event is significant and emotionally arousing.
You draw out the exact condition under which reliability may improve, rather than making a blanket claim. That conditional reasoning is what makes the argument nuanced.
Unlike Loftus and Palmer, this was a real-world event with genuine consequences for the witnesses, so the findings have greater ecological validity.
This comparative move is especially effective because you do not discuss the study in isolation. You actively weigh it against earlier evidence, which helps the essay build a genuine debate.
Yuille and Cutshall therefore qualify the claim that memory is unreliable.
This opening sentence keeps the paragraph tightly linked to the central argument. The verb 'qualify' is especially strong because it signals refinement rather than simple contradiction.
Because Yuille and Cutshall studied a naturally occurring event, many variables could not be controlled.
This is a relevant methodological detail that supports your caution about the study. You choose evidence that directly explains why the findings cannot be accepted uncritically.
Even so, the study is valuable because it shows that the reliability of memory depends on context rather than being uniformly poor.
This is strong evaluative synthesis within the paragraph. You acknowledge limitations but still explain the study’s contribution to the overall argument, which is a mature critical move.
Further support for the idea that memory is sometimes reliable comes from the distinction between central and peripheral details.
This is an effective signpost because it introduces a final layer of refinement to the argument. It shows the essay is still developing, not just repeating earlier points.
In emotionally significant situations, individuals may remember the central aspects of an event very well, even if peripheral details are less accurate.
This is a precise conceptual detail that extends the discussion beyond named studies. It adds psychological specificity to the claim that reliability can vary within the same memory.
Reliability is not absolute.
This brief sentence is powerful because it crystallises the essay’s central insight in clear, controlled language. The phrasing is concise and decisive, which gives the argument real authority.
Overall, memory is reliable only to a limited or moderate extent.
This conclusion returns directly to the question and gives a clear final judgement. It synthesises the discussion into a settled position rather than merely repeating earlier points.
Therefore, the best answer is not that memory is reliable or unreliable in general, but that its reliability is conditional.
This is an excellent concluding synthesis because it captures the essay’s whole line of reasoning in one balanced judgement. You show that the debate has been weighed, not just described.