The Fundamental Attribution Error Problem

Perhaps one of the biggest flaws in human thinking is the Fundamental Attribution Error.  The Fundamental Attribution Error (FAE) is tendency for people to put an excessive emphasis on a person’s internal characteristics, while being inconsiderate of the external factors of a situation.

To use a simple example, when something happens, we blame it on the person rather than the situation. Imagine if someone is in a hurry and does something rude – you may immediately attribute that to the person being rude. Perhaps they were in a hurry because they had an emergency or some justifiable reason. People do not consider the greater context of why they might have been rude, whether or not this was a usual occurrence for the person, and whether or not there were any extenuating circumstances.

Here is a more detailed explanation and if you are interested, yet another here from the Guardian.

Generally, we humans are prone to giving the FAE to other people, and seldom ourselves. When someone does something bad, likely to you, you are far more likely to attribute that to them being a bad person rather than the situation. By contrast, when you do something that is perceived negatively by others, you are far more likely to give yourself a break.  There is also the fact that when something bad happens to you, you tend to take it personally.  I think that this may be a sort of “egocentric effect” that goes on with all of us.

Why is this such a bad problem?

In everyday life, there is the tendency to take everything at face value and not look for alternative explanations. That can lead to incorrect and sometimes dangerous conclusions because we attributed the outcome disproportionately to the person without considering any other factors.  The consequences can be significant:

  1. We don’t actually find the real cause of a problem or success.
  2. It can contribute to prejudice against a certain racial group or animosity towards a person whose intentions may have been pure.  We may also have an exaggerated assessment of how a person behaves based on one incident.
  3. We may have inaccurate assessment’s of a person’s competence. We fail to consider the difficulty in achieving a problem. I will explain this one below.
  4. Where we make first impressions or our initial judgements about a person, a situation, or really anything, we may not correct them when the data does not suit our cognitive biases.
  5. It can disincentivize learning more about a situation or a person when we think that person or that situation behaves a certain way normally.

Obviously these are very real and very sobering consequences.

Research has indicated that even highly qualified professionals are not immune to this. In fact, it is so rooted in our psyche that people assume that a high growth group is automatically perceived to be more competent, regardless of the difficulty in achieving said task. The paper explores various situations, such as whether or not employers would choose a higher GPA major even in cases where a university was known for engaging in grade inflation. I will quote their conclusion:

Each of our studies supports the hypothesis that people rely heavily on nominal performance (such as GPA) as an indicator of success while failing to sufficiently take into account information about the distributions of performances from which it came. To the extent that admissions officers and hiring managers generally show the same biases we found, graduate programs and businesses are collectively choosing to select candidates who demonstrated their merit in favorable situations rather than selecting the best candidates. The consequences could be substantial for both the sufficiently qualified but unselected candidates as well as for the organizations that systematically select lower performing candidates than they could.

In other words, we are not very good at discounting easy successes, and acknowledging when people’s performance was good given the challenges that they faced.

What does this mean for individuals?

When we are successful, we like to attribute it to ourselves, that we are great people, versus something else.

This can lead to some dangerous confirmation biases, misunderstandings, and a sense of unjustified arrogance. In an article for Harvard Business Review, Francesca Gino and Gary P. Pisano argue that there are three very big problems that impede learning:

  1. The Fundamental Attribution Error: This makes us conclude that our own talents caused success.
  2. Overconfidence Bias: Due to the above, we become overconfident in our own abilities. While, the authors argue some confidence is good, an excess is bad.
  3. Failure to ask Why: When everything is going well, we tend to attribute this to our abilities and not anything else. We do not ask the all important “why” questions.

Gino and Pisano go into racing and argue that in the case of Ducati, they initially became more successful than anticipated, then they failed to investigate the correct reasons, and ultimately that year, their performance was disappointing compared to their high expectations.

The problem is that in any outcome in the real world, especially for complex problems, there are many interdependent factors that will impact the outcome. Success in particular can be very dangerous in that it leads to overconfidence and misunderstanding a situation. It does not lead to the necessary levels of reflection.  In some ways, success begins to work against the person that was successful and breeds a dangerous sense of complacency.

The authors recommended several solutions:

  1. Be happy if you are successful, but examine the why very closely.
  2. Do postmortem reviews, even for very successful projects. People don’t like to do them, but they are very important.
  3. Focus on the long-term time horizons (or the most appropriate given the time horizon).
  4. When things go well, we over-focus on how to replicate. Understand that although important, our ability to learn when we replicate is not always.
  5. Experiment a lot when things go well.

They argue the question that people should ask is not, if we are doing well, but what experimentation is being done.

A dangerous real world example

I have noticed in the real world, particularly amongst political conservatives in the US, but also elsewhere, their relentless attacks on the poor and marginalized groups. They fail to consider what sociologists called “life chances”. Political conservatism in the US has always argued that society is a meritocracy, and that the poor are that way due to their own poor choices, rather than other external factors, such as racism, a bad economic situation, lack of access to quality healthcare, vastly unequal access to education, and other advantages that the wealthy have.  It is a very Social Darwinistic view to be taking, but it reflects the sentiment of so many people. Compounding this is willful ignorance, which is another cognitive bias.

It is all too easy to dismiss an entire group’s poor economic performance as being due to being inferior, rather than considering other external factors. Ironically, the ideology advocated by the political left might make society far more meritocratic in that it would lead to a more even starting field for life chances.  That seems to be true from everything from education to even breastfeeding.   There does seem to be a negative correlation as well between income inequality and social mobility.

I am glad that this attitude is not nearly as prevalent here in Canada, although distressed that there are some (and I fear increasing numbers, particularly if inequality continues to rise) Canadians that seem to harbor similar views.

What does this mean for gadflies?

Well, seeing that research has indicated that even highly trained people are not immune to the fundamental attribution problem, we might start with how to overcome it.

Actually, I am worried that gadflies and intellectual oriented people may be even more vulnerable than normal people. I have heard the argument before that the very intelligent overvalue the value of intelligence and are thus more prone to the fundamental attribution error. In that case, intellectuals must be extra cautious about their own cognitive biases.

Leave a Comment

Your email address will not be published. Required fields are marked *