Why do we forget?

I realised earlier today that whilst I’ve written several posts about memory, for example this one, about the different types of memory, the link between smell and memory, whether our memory is trustworthy, and about those with perfect memory syndrome, I’ve never actually written a post about the opposite – forgetting. Why is it that we often can’t remember something so simple as what we had to eat yesterday, or a piece of information we need to know for an exam? Read on to find out more..


One theory is the Trace Decay Theory of forgetting. This assumes that memories leave a trace in the brain, and if we don’t activate this trace (by thinking about the memory) then it fades, or decays. This theory involves our short term memory, which has a limited duration and can only hold onto information for around 30 seconds. However, it is actually pretty hard to test, meaning there isn’t much evidence to support it. It also doesn’t explain why people can remember things even though they haven’t thought about them for years, which is at odds with trace decay theory.

An alternative theory involving the short term memory is Displacement Theory. This theory is based on evidence which has shown the capacity of the short term memory to be between 5 and 9 items (Miller, 1956). Once new information enters our short term memory, other items in there are displaced. This has been illustrated by asking participants to remember a list of words. Results of experiments using this method have found that people are more likely to remember the words at the beginning and at the end, the ones in the middle have been ‘displaced’.


Interference Theory explains forgetting in terms of our long term memory. Have you ever typed in your old password and wondered why it wasn’t working? That’s an example of proactive interference – old knowledge interfering with what we know now. Or how about if you’ve broken your new phone and have to go back to using your old one, but keep pressing the wrong buttons? That’s retroactive interference – new knowledge interfering with what you used to know. Anderson (2003) explains interference as a failure of inhibition in the brain, whilst it might be useful to forget some things over time (e.g. what you had for dinner 3 weeks ago), there are other things which we need to remember, despite new learning. A single retrieval cue (such as sitting at your computer) can link to more than one memory (your old and new password), meaning the correct memory needs to be selected. However a problem with this mechanism means that as well as forgetting potentially distracting memories, problems with inhibiting other memories triggered by the same cue means that useful things are forgotten too.

The above theories assume that the memory has been forgotten because it no longer exists. But what if the problem isn’t with the memory itself, but the process of remembering known as retrieval? Retrieval failure happens when the memory is still contained in our long term memory, but we are unable to access it because certain cues are not there. These cues can be anything such as context about where you were when you learnt the information (external), or how you were feeling (internal). Goddon & Baddeley (1975) asked a group of divers to take part in a memory experiment. Half learnt a word list on land, and half underwater. Half of the group who learnt the list on land then had to recall the list on land, whilst the other half had to do this task underwater. The same happened to the participants in the underwater learning group. They found that participants who had to recall the words in the same setting as they learnt them in performed significantly better than those whose context had changed.

What about when forgetting is more serious? Amnesia is more severe than the types of forgetting we experience in day to day life, as it can involve forgetting large proportions of previous life events or information and is often caused by trauma to the brain. Perhaps the most famous case of amnesia was in Patient H.M., who had most of his hippocampus (structure in the centre of the brain which is thought to be responsible for long term memory) removed to cure his severe epilepsy. Whilst successful in reducing his seizures, he was left unable to retain any new information for more than a few minutes. If you’d like to read more about what H.M.’s case taught us about human memory, I’ve also written a post about that here.



Anderson, M.C., 2003. Rethinking interference theory: Executive control and the mechanisms of forgetting. Journal of memory and language49(4), pp.415-445.

Godden, D.R. and Baddeley, A.D., 1975. Context‐dependent memory in two natural environments: On land and underwater. British Journal of psychology66(3), pp.325-331.

Miller, G. (1956). The magical number seven, plus or minus two: Some limits on our capacity for processing information. The psychological review, 63, 81-97.

Scoville WB, Milner B. J. 1957. Neurol. Neurosurg. Psychiatry. 20:11–21



The Nocebo Effect

The Placebo Effect: a psychological effect in which a treatment which contains no active medical substance causes an improvement in symptoms. For example, a participant in a trial takes a sugar pill believing it could be real medication and find their back pain goes away.

This effect has been well documented and is relatively well known (for more information read my blog post here). But what about the Nocebo effect? In this instance, the opposite happens. A participant in a trial takes a sugar pill, or receives a fake injection, and start to feel negative side effects of the medication. How is this possible when no active medication was received?

To begin to explain how this effect occurs, I’ll start by telling you a bit about how clinical trials are conducted. Before a participant consents to take part in the study, they have to read a participant information sheet which explains all the details of a trial and what will happen. In a drug trial, such as one testing a new medication to help persistent back pain, participants will also have to read a list of potential side effects, much like those you find on the leaflet that comes in the box with medication. When participants sign up, they are told that they might receive the real medication, or they might receive a sugar pill. Having the control group of those who receive fake medication is important in clinical trials, as it allows you to show that any improvement is due to the medication being tested and not other factors such as symptoms improving over time. The reason control groups are given fake medication instead of having no medication at all allows for researchers to see how much of the improvement of the real medication is due to it’s active ingredients, and to show that participants haven’t just got better because of the placebo effect.

image from http://www.thehealthsite.com/diseases-conditions/mind-blowing-facts-about-the-nocebo-effect-k0517/

Even though participants have received the placebo medication they can still believe it is the real one – a placebo should be administered in exactly the same way as the real medication to be a true control. Therefore, it is this belief that they have taken the real drugs that can lead them to report side effects from it. One review of the evidence shows that around a quarter of participants taking a placebo drug experience adverse side effects from it, and that this can be higher than the participants taking the real medication! (Barksy et al, 2002). Visual cues can also induce nocebo effects: one study tested how participants rated the effectiveness and side effects of either branded or unbranded drugs (both in fact were placebos). Perhaps unsurprisingly, participants rated the branded drugs as more effective, and thought the unbranded drugs caused more side effects (Faasse et al 2013).

Several explanations have been put forward to explain the nocebo effect, including conditioned responses or participant’s expectations. For example, a doctor giving you an injection warns you that it might hurt, so you feel subjectively more pain than if they had been reassuring. Some studies investigating the neural basis of the nocebo effect in pain have hypothesised that the effect is caused by increased activity in certain areas of the brain such as the hippocampal network (which is involved in pain modulation) (Ploghaus et al, 2001). This activity is in turn caused by increased anxiety, brought on by the expectation of pain.

This brings a certain ethical dilemma for healthcare professionals and those running clinical trials. It is important the the patient or participant is given all of the information, in order to give informed consent. However, if giving someone more information would cause them to feel more pain, what would you do?



Barsky, A.J., Saintfort, R., Rogers, M.P. and Borus, J.F., 2002. Nonspecific medication side effects and the nocebo phenomenon. Jama287(5), pp.622-627.

Faasse, K., Cundy, T., Gamble, G. and Petrie, K.J., 2013. The effect of an apparent change to a branded or generic medication on drug effectiveness and side effects. Psychosomatic medicine75(1), pp.90-96.

Ploghaus, A., Narain, C., Beckmann, C.F., Clare, S., Bantick, S., Wise, R., Matthews, P.M., Rawlins, J.N.P. and Tracey, I., 2001. Exacerbation of pain by anxiety is associated with activity in a hippocampal network. Journal of Neuroscience21(24), pp.9896-9903.


Epilepsy is a neurological condition characterised by repeated seizures. Seizures are caused by electrical activity in the brain, although may appear differently from person to person (not all seizures involve convulsions, despite what you might think).

As with many conditions there is not a single cause that can be identified as a precursor to epilepsy. Genetics (a mutation in the KCNC1 gene has recently been identified as a cause of a progressive inherited form of epilepsy – Muona et al 2015), brain tumours, or head injuries, and the cause of many patients’ epilepsy remains unknown. Several studies have shown that you are more likely to develop epilepsy after a head injury e.g. Christensen et al (2009) found that people were 2% more likely to develop epilepsy after a mild head injury. This rose to 7% more likely following a severe head injury, with risk also increasing slightly with age.

The image below is taken from the EFEPA and shows what to do if someone is having a seizure:


As mentioned earlier there are different types of epileptic seizures which depends on which part of the brain they originate in. Seizures can be classified by how much of the brain is affected: partial/focal seizures (when only a small part of the brain is affected) or generalised (if most of the brain, or all of it, it affected).

Focal seizures can also originate in different parts of the brain, with the temporal lobe being the most comment (epilepsy.com). The temporal lobe is the part of the brain above your ear, and is responsible for processing hearing, and our memories (this is simplified – it does a bit more than this!). Therefore, one of the common features of temporal lobe epilepsy is memory disturbances (Ko et al, 2013). The famous patient H.M.’s amnesia was caused by an operation to remove the source of his severe temporal epilepsy – this was carried out in the 50s before brain functions were accurately known and too much of the medial temporal lobe was taken away. This destroyed part of the hippocampus, the structure in the brain responsible for memory processing. Due to the nature of his amnesia, he was probably one of the most studied individuals ever in psychology. See this post for more on H.M. and memory research. Operations are carried out to remove part of the temporal lobe in patients now with much better outcomes!

The second most common is frontal lobe epilepsy, where seizures originate in the front part of the brain. They often occur during sleep, and can affect the motor areas of the brain, leading to problems with motor skills (e.g. Beleza & Pinho, 2011). If patients are not eligible for surgery to remove the specific part of the brain responsible for the seizures, anti-convulsive medication and electrical brain stimulation can be helpful in reducing symptoms (Kellinghaus & Luders, 2004).




Smell and Memory

I’m sure this has happened to you before – you’re walking down the street and you smell something that takes you back to a holiday, or a time when you were younger. It could be the smell of a sweet shop or someone’s perfume, and you are taken straight back to a moment from years ago. But why are smells so linked to memories?

A simple answer is that this link is due to how the brain is organised. Our sense of smell is triggered by a molecule that enters our nose and binds to the hair-like projections (cilia) on neurons at the top of your nasal passage. These neurons project to a part of the brain called the olfactory bulb, which run along the front of the brain, at the bottom. This structure is thought to be involved in interpreting these signals and processing information about smells.

What’s interesting about the olfactory bulb is that it’s the one part of the brain responsible for our senses that has projections to and from the areas of our brain responsible for memory and emotion – the hippocampus and amygdala. You can see this from the image below:



This explains why smells can trigger memories and emotions. The hippocampus is responsible for our episodic memories in particular – personal memories about our lives, which is why it is this type of memory activated by smell. One theory about why these connections exist between the hippocampus and the olfactory bulb is that they enable us to recognise smells from previous experience.

Studies have shown that using smells to trigger memories can be more effective than cuing them with words. For example, Maylor et al (2002) asked young and old adults to recall autobiographical memories associated with 6 cue words. They were then shown the same words and were asked to recall new memories, and for half of these words the appropriate smell was presented too. The researchers found that for both age groups, the participants recalled twice as many memories when the smell was presented too, showing the large impact of smell and memory recall.


Seeing Faces

It has been argued that faces are a special type of stimuli, which we are able to process easier than other items in the environment. For example, young infants prefer to look at features forming a face, rather than scrambled features (e.g., Johnson & Morton, 1991), which suggests that we have this enhanced ability to process faces from birth. Thomas (1980) argued that faces are processed more holistically than other objects. But how are faces processed in the brain?

One hypothesis is that there is a special area in the brain for processing faces – the Fusiform Face Area located in the fusiform gyrus (shown below). This was named by Kanwisher (1997) in a fMRI study which found that this area was more active in participants when they viewed faces, rather than other stimuli. It also showed more activity for whole faces, rather than scrambled ones. They therefore concluded that this cortical area is specialised for processing faces.


Further evidence for this area being specialised for face perception is shown by patients with prosopagnosia – an inability to recognise faces. One of the causes of this disorder is damage to the FFA and surrounding cortical areas, which suggests this area is important for normal face perception.

However, although this area might be important for face perception, there is evidence from patients with damage to the FFA who have no trouble detecting faces (e.g. Tranel et al, 1998). Therefore, the FFA might not be necessary for face perception. Some studies have also suggested that the FFA alone isn’t sufficient for normal face perception. Behrmann (2003) found that patients who had prosopagnosia from birth (not from neurological damage) had intact FFA and normal activity there. This evidence suggests that things might be a bit more complicated than simply having one area which is responsible for face perception.

One alternative explanation is that there is a network of cortical areas, which all interact to process faces (Haxby, 2000). This study used a pattern analysis method and found that several other areas in the brain were active when participants were shown pictures of faces e.g. the superior temporal sulcus and lateral occipital gyri, suggesting the theory of a single area such as the FFA is over-simplified.

I hope you enjoyed this post, and thank you for reading.