How reliable is our memory under conditions of stress?

In an earlier entry about false confessions, I briefly touched upon the topic of how memory works. The way it works barely resembles the way artificial memory systems work. Rather than accurately retrieving information, we reconstruct our memories every time we recall them.

The topic of memory, just like many other Psychology topics, is subjected to a lot of misconceptions and myths. The way we collect, store and recall memories is extremely complex with different factors like emotional activation or information type playing an important role.

How do emotions affect our memories?

It is a widespread idea that emotions have a strong effect on memory. As early as the inception of psychoanalysis, its proponents theorized that memories of traumatic experiences or emotionally disturbing events tended to be repressed. Despite the fact that repressed memories are still a popular concept, the actual incidence of repressed memories is very small. So much so, that many psychologists believe they should not be considered a real phenomenon.

Oddly enough, the belief that emotionally charged events are more memorable is also a common idea. So, what truth is there to these popular assumptions?


There is considerable scientific evidence that proves that information with emotional content is remembered more vividly than neutral information. [1] This happens not only with autobiographical memories – important events in our lives- but also with information provided in a lab in the course of an experiment.

Emotional information is remembered more accurately because of two special features that make its initial encoding more effective; valence and arousal. Emotional information, as opposed to neutral one, is evaluated by us as either positive or negative (valence). Emotional information also provides some type of physiological and psychological activation; having a soothing or agitating effect on us (arousal). Information is more effectively encoded when processed in this context of activation and interpretation as it helps us to combine new information with already existing memories.

Unreliable accounts

If emotionally charged memories seem to be remembered better than neutral ones, why are people so unreliable when retelling an emotional story?

Even though emotional events are generally remembered better than neutral ones, very intense emotional activation can make people forget details. Intense activation leads to attention narrowing which makes us focus on the central aspects of an event, ignoring the secondary details.

The Yerkes–Dodson curve; a classic, empirical law that dictates that performance increases with physiological or mental arousal, but only up to a point, can explain the difference in accuracy of memories on subjects under stress. When a certain point of activation is reached, performance decreases.

Yerkes-Dodson curve

The case of eyewitness testimony

The study of memory accuracy is of extreme importance to the field of Forensic Psychology. Eyewitness testimony plays a vital role in countless criminal investigations and court proceedings but the question that I get often times is; how reliable is eyewitness testimony?

Regarding eyewitnesses to a crime or accident, we should understand that they can vividly remember a given event and its most important features but; depending on their physiological activation, they won’t be able to give specific details about secondary aspects. Collateral information like the physical appearance of an aggressor or the color and size of certain objects won’t be stored accurately.

It seems to be the case that right after experiencing a traumatic event, only essential information is remembered. In the hours or days following the event, some secondary details and collateral information can be recovered and remembered.

The problem comes when an overzealous investigator questions a witness, providing details with their questions, which the subject can then incorporate into their own recollection, leading to inaccurate testimonies.

So in short, the accuracy of memories in the context of stressful events depends on:

  • The level of activation of the subject.
  • The amount of time that has elapsed since the event.
  • How the person is being questioned.

What could be done about this?

Interrogation techniques need to be refined. Not only police investigators need to formulate neutral and simple questions that don’t provide any information or opinion regarding the case. Also, the questioning by lawyers and prosecutors can alter a testimony, so they should be conducted carefully as well. When it comes to recognizing suspects, police lineups should be conducted similarly to how a scientific investigation, making use of double blinds and preventing anyone from giving any hints to the person doing the identification.

[1] Remembering emotional events. A. Burke (1992)


The hidden dangers of repeated head trauma

Earlier this year, Aaron Hernandez, an American football player, killed himself in his cell after being tried for a double murder. He was a rising football star when his legal issues first became public and many lamented how this put an end to what could have been an extraordinary career. Hernandez’s family decided to donate his brain to the CTE research center in Boston, with the intention to discover whether the athlete was affected by this neurological condition.

Although I have heard about Chronic traumatic encephalopathy before, it wasn’t until I read the news on Aaron Hernandez that I decided to delve into it. CTE is barely mentioned outside of the United States and I think that European sports authorities should be aware of the problem of repeated head trauma in players.


The consequences of repeated brain trauma amongst boxers are well documented. Dementia pugilistica -a subtype of CTE- has been studied since the late 1920s. Repeated head injuries in boxers can cause a pattern of neurodegeneration that produces dementia-like symptoms; cognitive and motor impairment, as well as mood and personality changes, which range in severity.

For many years, it was thought to only affect boxers, leading some medical professionals to ask for a ban on the sport. It is a recent discovery that repeated mild traumatic brain injuries increase the risk of neurodegenerative disease in athletes who play contact sports.

Although most research has been carried out in the context of American football, it is important to note that there are cases of CTE in many other sports; hockey, rugby, baseball, martial arts, soccer and motocross (BMX), as well as in people with military service.

Chronic traumatic encephalopathy usually manifests itself after about a decade of repeated injuries. Considering that many athletes begin their sports practice early on (ages 12 to 18), first-stage symptoms can be observed in young players in their early twenties [1]. Nevertheless, problems usually begin later in life.


CTE shouldn’t be confused with a post-concussion syndrome where symptoms are present shortly after a concussion but eventually go away. CTE consists in a progressive degeneration of the cortex and neuronal loss that, once it has been triggered, won’t stop in the individual’s lifetime. It has many similarities with Alzheimer’s and Parkinson’s disease, but in the case of CTE, we do know it has a clear environmental cause.

The diagnosis can only be confirmed post-mortem, once brain material is available to be examined. The research center I mentioned at the beginning of the article has examined the brain tissue of 94 former NFL players that presented symptoms and has concluded that 90 of them suffered from CTE. [2] Some of them had committed suicide at an early age, just like Aaron Hernandez did.

So, what can be done about this?

Since CTE is caused by recurrent concussions, the only way to prevent it, it’s by reducing the amount of exposure to head trauma. Unfortunately, that’s not an easy task. Sports-specific helmets seem to reduce injury in some sports. It is important to pay attention to the design and materials of the helmet as well as to make sure that it is replaced once it has taken a big hit.

Another very important factor is the “Return-to-play” criteria after a head injury. Studies suggest that a head injury should require at least 4 to 6 weeks of recovery time to avoid re-injury. Sadly, in some cases, these guidelines are not strictly enforced.

Lastly, athletes would benefit from some sports rules changes as well as from severe penalization for intentional hits to the head.


[1] Chronic Traumatic Encephalopathy in Athletes: Progressive Tauopathy following Repetitive Head Injury

[2] CTE Research Center Case Studies

The Psychology of False Confessions

Confessions have always been regarded as the best type of evidence that can be presented in a court of law. Centuries ago, a confession amounted to a conviction. Nowadays, the majority of criminal cases are solved by virtue of a confession, and jurors give a great importance to confession statements even when they are later retracted. The recent boom of true crime -with documentaries like Making a Murderer, Shadow of Truth or podcasts like Actual Innocence– has put the topic of false confessions in the spotlight.

It is very hard to assess the incidence of false confessions in a given country. The Innocent Project estimates that 25% of exonerees in the United States had confessed to the crime. Today I will like to take a look at the Psychology of false confessions, how and why do people confess to a crime they haven’t committed.

Types of false confessions

Before I begin, I will quickly touch on two types of false confessions that don’t need a lot of explanation. Firstly, we have the voluntary false confession. Voluntary confessions are common in very high-profile cases where someone confesses to a crime hoping to gain fame or recognition. Another type of voluntary false confession occurs when the person confessing is trying to protect the actual perpetrator. For instance, a mother trying to prevent her son from going to jail for killing her abusive husband.

The second type would be the coerced compliant confession. This confession is made by someone who has been forced to confess. The confession can be forced by means of physical violence or threats of being charged with other crimes if they don’t confess to the crime in question. It is important to note that in this instance, the subject is completely aware that they did not commit the crime and are just “cooperating” to avoid an adverse situation.

The third type of false confession is the most interesting one from a psychological point of view but also very unsettling to consider. It is called the coerced internalized confession. In this case, the person confessing to a crime they didn’t participate in, will eventually believe they committed such crime and their memory of the event in question can be altered forever.


How does a coerced internalized confession occur?

To understand this, we need to understand how human memory works. Human memory doesn’t work like computer memory, our memories are being transformed constantly. Every time we recall a stored memory, we reprocess it again and the end product -that we send back to storage- it’s something very similar to the original memory but with a few modifications. We introduce these modifications ourselves when we think about how that event unfolded. However, accounts from other people or information added by others can also change the way we remember said event.

The good news is that it takes a “perfect storm” of circumstances for a coerced internalized confession to happen, the bad news is that the most vulnerable people in our society are at risk.

The first factor we need to consider is how interrogations are carried out. Police interrogations are designed to deal with the “typical” criminal personality; extrovert, cunning, and extremely suspicious of law enforcement. When the same method is used on the opposite type of personality -introvert, gullible, and trusting of the police force- the suspect can feel easily overwhelmed and unable to cope with the stress of the interrogation.

Interrogations are usually long and conducted in a hostile environment with little sensory stimulation, no windows and only having contact with the investigators. The close, social interaction with the investigator, the uncertainty about one’s own safety and immediate situation as well as being confronted by an authority figure, make the suspect a very suggestible subject.

Investigators are allowed to present false evidence in the course of an interrogation. They can tell the accused they have DNA evidence, positive eyewitness identification or accomplice confessions. All this can be used as they please in the course of the interrogation in order to build their case. They are also allowed to use psychological manipulation; being on their side, minimizing the crime and its consequences or, conversely, maximizing them.

Suspects that finally succumb to these interrogation tactics and give a false confession have certain characteristics. Victims of false confessions are usually vulnerable people. Be it because of mental disability, low IQ, drug or alcohol use as well as withdrawal, fatigue or stress, they are put in a position where their memories are easily moldable. A low self-esteem, a lack of assertiveness as well as typically poor memory can also be determinant.

As the length of the interrogation grows, the suspect focus changes from long-term consequences (avoiding going to jail) to short-term consequences (being allowed to leave the stressful situation, resting or seeing and talking to their loved ones) so a coerced-internalized false confession is made. At this point, the highly suggestible subject has also started doubting that what he remembers is true and starts to believe the investigators’ version of events. This version has been repeated and reworked so many times that the accused has made it their own.

Final thoughts

The field of human memory research and false memories is fascinating one but I will not go into more depth today. I would like to see law enforcement being trained more adequately for suspect interrogations. A stronger focus should be placed on identifying and understanding a suspect’s shortcomings (like mental disability) in order to adapt their interrogation method.

It is clear that law enforcement methodology helps to solve many cases but in the cases in which their methodology elicits a false confession, the consequences are catastrophic. It is also important that we inform ourselves about the topic and in the case of ever serving on a jury, we should require more evidence than a confession to convict an accused person.

Source and recommended reading:

Gudjonsson, G. (2003). The psychology of interrogations and confessions.

The problem with cognitive biases

I have already talked about how our minds make use of heuristics to solve problems when faced with cognitive limitations. These heuristics are shortcuts that are bound to be wrong sometimes. The 1970s marked the beginning of the heuristic and biases tradition which took over the study of human decision-making processes.

Currently, cognitive biases are being questioned by many researchers. Some believe they are a vague, overrated concept that unfairly dominates the field of reasoning studies. In order to explain this position, I will take a look at the confirmation bias today.

Confirmation bias

The notion of confirmation bias is simple to understand. In 1960, cognitive psychologist P.C. Wason coined the term [1]. He used it to describe how when we search for information we tend to favor the one that confirms our preconceptions, hypotheses or personal beliefs. It is also known as myside bias and affects inductive reasoning (the type of reasoning in which the premises would lead to a probable solution, not a certain one).

The concept of confirmation bias has expanded and is supposed to affect not only to the task of information search but also to interpretation and recalling. In fact, confirmation bias has become an umbrella term that covers any instance in which beliefs or preconceptions influence information processing.


Confirmation bias is seemingly unrelated to intelligence [2][3] and has been theorized that confidence levels might be at play. Individuals with low confidence levels might tend to avoid information that contradicts their beliefs while those with higher confidence levels tend to seek out antagonistic opinions to form their arguments.

Examples of confirmation bias

Considering the preceding definition, the instances in which we incur this cognitive bias are endless. From politics to the belief in pseudoscience, to rushed medical diagnosis, to police investigations, to almost any everyday social interaction; confirmation bias could be affecting us constantly.

It is quite common to disregard any evidence that doesn’t sustain our political point of view by saying that it is flawed or inaccurate. It is just as common that scientific innovations or rare experimental results are heavily criticized by the scientific community.

Someone who believes in paranormal psychic activity is predisposed to remember vividly every time that event A was followed by event B. This same person will ignore or even forget all those instances in which event A wasn’t followed by event B.

We can find countless accounts of myside bias. But the question is, are they the result of a systematic cognitive bias that affects human reasoning equally or is it something else?

The problem with cognitive biases

As I said before, cognitive biases are not universal. They are dependent on individual differences – intelligence, previous training or certain personality traits. Training to solve experimental tasks used in cognitive bias studies as well as simply knowing about the existence of cognitive biases makes us less likely to fall prey to them.

Moreover, most studies that support their existence feature highly subjective tasks. In the case of confirmation bias, for instance, there is little evidence that it affects estimations of numerical results. This is the reason why I think we should be using the concept of motivated reasoning instead of confirmation bias.

I think it is problematic that the effect of cognitive biases cannot be identified across a wide range of contexts, seems to be task-specific as well as restricted to some people. As a result, we cannot affirm that cognitive biases are systematic patterns of irrationality that govern human judgment.

[1] Wason, P. C. (1960). On the failure to eliminate hypotheses in a conceptual task. Quarterly
Journal of Experimental Psychology, 12, 129–140.
[2] Stanovich, K.E. & West, R.F. (2007). Natural Myside Bias is Independent of Cognitive Ability. Thinking and Reasoning, 13 (3), 225-247.
[3] Stanovich, K.E.; West, R.F.; & Toplak, M.E. (2013). Myside Bias, Rational Thinking, and Intelligence. Association for Psychological Science, 22 (4), 259-264.

Recommended reading:
What Does It Mean to be Biased: Motivated Reasoning and Rationality by Ulrike Hahn and Adam J.L. Harris.

The Psychology of Emotions

Emotions are, alongside personality, the psychology topic, we more often refer to in our daily lives. We talk about how we feel -or felt- in a certain situation and how those emotions influence our conduct and choices.

Experimental research in the field of psychology and neuroscience has provided us with a deeper understanding of emotions and how they affect us. The general agreement among researchers is that there are six basic emotions: happiness, sadness, anger, disgust, fear, and surprise. They are considered our basic (or primary) emotions because they seem to be present cross-culturally. Another important feature of basic emotions is that each of them is associated with a specific pattern of physiological activity and facial expression.

It seems obvious that we experience a wider range of emotions than those basic ones. Complex or secondary emotions -like annoyance or envy- are made up of basic emotions and probably came about, and stayed with us, by way of cultural conditioning and association.

Psychologist Robert Plutchik proposes a model with eight primary and contrasting emotions.
The function of emotions

Emotion is undoubtedly a defining aspect of humans and animals, and a factor that (still) differentiate us from machines. Emotions also have a profound effect on our thoughts and actions which can lead to both a positive and negative outcomes.

So, apart from making us what we are, do emotions serve us a purpose? Yes, they do. Emotions have an important adaptive role as they get us ready for action when it’s most needed. A good example of this would be fear. We experience fear when we perceive a danger or imminent threat. In this case, the flight-or-fight response is triggered making us more alert, shifting our attention to negative stimuli and preparing the body -increased muscle tension, blood flow- to fight or escape.

However, a lengthy state of fear will have negative consequences. If a person is worried about losing their job and experiences it with fear and intense anxiety, the emotional response will probably be prolonged in time. Such person won’t be able to focus on their daily tasks and will have trouble sleeping at night because of this emotional activation.

Emotional biases

Emotional states can cause cognitive and memory biases. A clear example is how anger clouds our judgment and keeps us from making rational decisions. When it comes to processing information, we prioritize the emotionally-charged stimuli over neutral ones. Emotions can also alter the content of our memories as well as how or when a memory is recalled. Memories consistent with our current emotional state are more easily accessible.

The effect of emotional states on cognition and memory is a broad field of research. It is also of vital importance to clinical psychologists as it helps us understand how the minds of those battling anxiety or depression work.

Human emotion is fundamental to the study and treatment of mental disorders as well as to the understanding of human cognition and behavior.


George moved to the city after completing his Communication studies in hopes to break into the fashion industry. He wanted to be an editor in a big-name fashion magazine or website. George had been involved in fashion his entire life, modeling since he was a teen and keeping a blog where he documented his outfits and purchases.

George soon learned that his background was far from unique and that, without good contacts in the industry, even finding an internship can be a difficult task. He could have tried applying to lesser known fashion media outlets but that was not for him. George would either work for an influential publication or he wouldn’t work in fashion at all.

When I met him, he had just joined the company I worked for and he was being trained to become an account manager. He was quick to tell us about his previous modeling career, his aspirations to become a fashion editor and how this job was only temporary for him.


Two months in, he was frustrated by everyone. He believed that, even as a junior manager, he was bringing more to the company than most people. Account management was the hardest and most vital part of the business and he wasn’t being recognized enough for it. He also disagreed about crucial matters with his superiors.

George told me this while we sat at the best table of a fancy restaurant during our lunch break. He had insisted on being placed there and had been offended by where the waiter wanted us to seat initially. It was then that I realized how much certain things mattered to George. He needed to have the best of everything: the best clothes, the newest phone, the ideally located apartment and the fanciest vacation. He also needed to be constantly recognized and reinforced for being exceptional.

After a while, I realized that George would only talk to me when he needed to vent and discuss his frustrations. He had no interest in my life, but that was his modus operandi when it came to friendships. George had an amateur photographer friend who he would only call up to get pictures for his blog, another friend he would only meet up with to go to certain clubs where she was guest-listed, an office buddy from IT he would only talk to if he needed a personal tutorial about how to use certain software but ignored otherwise…and the list goes on. The more I listened to his stories, the more I realized he never did anything for anyone in return. When complaining about co-workers, he lacked empathy for other people’s problems and circumstances.

Intensive networking produced results and George got a new job opportunity in the industry he loved. But after a year working for a fashion magazine, he was frustrated with his career not going the way he had expected. He decided to quite some time after and borrowing some money from his parents, started his own online shop selling up-and-coming fashion designer clothes.

I only spoke with him a couple of times after this. He told me his business was going very well and that a big fashion retailer would buy them out soon. He was wondering whether he should stay in the company or take the money and start something new.

In reality, the offer never materialized and George kept running his company with moderate success. His online store was profitable but he felt like a failure. He also felt that a lot of people have let him down. He started seeing a therapist that diagnosed him with depression and narcissistic personality disorder.

Psychology and Machine Learning

Machine learning is quite the buzzword of 2017, as a subfield of artificial intelligence (another overused term), it has been around for decades. I remember being introduced to it by an enthusiastic professor in 2008 that told us how it was revolutionizing image analysis and the study of visual perception.

The boom of machine learning, which started when we gained the ability to work with big volumes of data, has been fueled by its versatility. Countless fields of work; financial services, fraud detection, logistics, medical diagnosis, natural language processing, marketing, and sales, are already benefiting from analyzing data through machine learning techniques. We are now starting to see the first applications of Machine Learning to Psychology problems.

Suicide prevention and machine learning

Earlier this year, a group of researchers from Florida State and Vanderbilt universities presented a study wherein a prediction model was developed to accurately identify the risk of attempting suicide in general and psychiatric patient population. [1]

The model can predict the risk of a suicide attempt with an accuracy of 80 percent (two years prior) and 84 percent (one week prior). In the general patient population, the accuracy is slightly higher. The false negatives are also lower than usual (from 1.2 to 3.5 percent). It is important to note that little progress had been made in the study of suicide prediction after decades of research.

The model

Since the study has not yet been published, I will quote an article written by Paul Govern on the Vanderbilt website [2] detailing the development of the model:

“[Researchers] started with de-identified records of adult patients seen at Vanderbilt from 1998 to 2005. They found 5,167 patients with billing codes indicating self-injury. A pair of clinical experts undertook separate reviews of this set, finding 3,250 cases, that is, 3,250 patients with a history of attempted suicide, and 1,917 controls, or patients with a history of nonsuicidal self-injury.

The de-identified records were pared down to demographics, diagnoses (…), socioeconomic status (…), health care utilization and medication information. To find predictors within these data, a machine learning technique called “random decision forests” shuffled this set of records repeatedly, each time building a “decision tree” upon comparing the shuffled set to the expert-ordered set’s strict segregation of cases and controls.

After thousands of shuffles, the algorithm became expert at predicting whether a randomly selected record from the training set was a case or a control. Finally, with a method called bootstrapping, the team used their training set to synthesize new data sets with which to measure the performance of their predictive models.

The second round of testing was set in the general patient population, using an additional 13,000 de-identified records as controls.”

As always, these results need to be taken carefully. The model identifies a combination of factors in the electronic records that could most accurately predict a future suicide attempt. Researchers are now waiting to see how useful the model is, once it is put to the test. The idea is to use it similarly as physicians use a cardiovascular risk score.


[1] Walsh, C.G., Ribeiro, J.D., & Franklin, J.C. (in press). Predicting risk of suicide attempts over time through machine learning. Clinical Psychological Science.

[2] Investigators use machine learning to predict suicide risk by Paul Govern