Home > Attention, Cognitive Bias > Don’t Worry, Your New Friend Isn’t Actually Following You

Don’t Worry, Your New Friend Isn’t Actually Following You

Imagine you are a college student at a party on a Saturday night. A friend introduces you to a guy that you have never met before; in fact, you have never even seen him before. The next day, you see the guy you just met in the dining hall, and then again later that afternoon in the library. Over the next few weeks, you start to feel like you see this guy everywhere you go on campus. This is called the frequency illusion. You may think that you are seeing him more often, but this is a distortion of reality and likely false.

When you meet a new person on campus and then you start seeing them all the time.

So, why are you feeling this way? It’s due to the frequency illusion, which is a cognitive bias that describes our tendency to notice something we have recently been introduced to much more often than we remember in the past. You may begin to notice the boy from the party more often, even though you do not recall ever seeing him before. The two major cognitive aspects of the frequency illusion are confirmation bias and selective attention (Zwicky, 2006). Confirmation bias occurs when people actively seek ways to confirm their original beliefs, while selective attention refers to our ability to focus on a particular stimulus while in the presence of multiple stimuli. Since attention is a limited resource, we are not able to attend to all of the stimuli that may be present in our environment. We need to recognize which is the most relevant, and dedicate our attentional resources to that stimulus. The two combine to create the frequency illusion; from the example above, now that you have met this guy, you choose to pay attention to him when you encounter him on campus (selective attention). Once you see him around a few times, you believe that he is everywhere and start to look for ways to confirm this belief (confirmation bias). 


We only see the objective facts that confirm our original beliefs.

Mendel et al. (2011) performed a study that experimentally demonstrated confirmation bias in the realm of medical misdiagnoses. Psychiatrists were presented with an initial case and asked to make a diagnosis. They were then given the opportunity to search for more information about the patient and make a final diagnosis. One out of eight physicians and one out of four medical students demonstrated confirmation bias when searching for additional information about the patient, looking to confirm their original diagnosis. So, what does this tell us about confirmation bias? We tend to seek information that will confirm our initial belief, which also means ignoring information that may contradict this belief. When we think that we are seeing something more often soon after it is brought to our attention, we may find ourselves actively searching for this novelty. In the example from the introduction, we start to look for this new person around campus to confirm our belief that we are indeed seeing them more often now that we have been introduced. Then, once we have confirmed our belief, we have fallen victim to the frequency illusion.

Selective attention also plays a key role in the frequency illusion. In the dichotic listening task performed by Grey and Wedderburn in 1960, participants are presented with alternating words and digits across ears. For example, the participant may be presented with mice, 1, cheese in one ear and 4, eat, 7 in the other. It is expected that participants will report the information by ear. In this case, participants often reported “mice eat cheese” which suggests that meaning forces attention to shift across ears. This is in alignment with Treisman’s Filter Attenuation Model, which illustrates attention as a permeable filter that prevents irrelevant information from being processed, but allows information that is relevant to the individual or the context to be processed.

Trainman’s Filter Attenuation Model

This means that information that we selectively attend to will be processed for meaning and transferred to short term memory, but stimuli that are particularly important to you, such as your name, or relevant based on the context, will have a low threshold. Information with a low threshold requires less cognitive resources to become activated. Therefore, this information will also be processed for meaning. This study helped to discredit the formerly accepted bottleneck model developed by Broadbent, in which only attended information would be processed for meaning, and all unattended information would be filtered out and left unprocessed (Cowan et al. 2001). In the previously mentioned dichotic listening task, “eat” has a low threshold due to the context of “mice” and “cheese”, even though it is in the opposite ear. Although information in this task is usually reported by ear, context brings the proceeding words closer to the threshold, which allows them to be processed for meaning, indicating that the attention filter is indeed permeable, allowing relevant information to get through. Before we are introduced to new information, it is still present, we are just not aware of it and it is not relevant. However, once we are made aware of the information, it becomes more relevant to us and we are able to selectively attend to it and notice it much more often. In the example from the introduction, the person that we just met is not necessarily new to campus, we just had never noticed them before. But now that we know them, their presence is more relevant to us and we are more aware of them, leading us to believe in the frequency illusion, feeling as though we see them more often than before.

Now that we understand confirmation bias and selective attention, how do they combine to form the frequency illusion? Galbraith and Underwood demonstrated the frequency illusion in a study performed in 1973. Participants were presented with concrete and abstract words of equal frequency. Participants tended to perceive that there were more abstract words than concrete words on the list. This is due to the fact that the abstract words were new and unfamiliar, which captured the participant’s attention. Participants reported seeing the abstract words more often than the concrete words due to the frequency illusion. Participants selectively attended to the abstract words, and then searched for them to confirm their initial belief that they occur at a higher frequency.

Wait, didn’t I just see you?

So how does the frequency illusion effect our everyday lives? This cognitive bias can make us question ourselves, and in some cases, feel as though we are going seeing things that are not actually present. In the example from the introduction, when you meet someone new, you start to feel like they are following you everywhere. However, you are most likely not running into the person any more often than you had in the past, you are just noticing it more now that you know them. The frequency illusion can also play an influential role in advertising. Say, for example, a friend tells you about some new clothes they recently bought at a store you have never heard of before. Now that you have been introduced to this store, you may start to notice billboards, television commercials, and other forms of advertising for the store. Since the advertisements are suddenly capturing your attention, you are more likely to think the store must be increasing in popularity, and want to check out the store for yourself. Another common example of the frequency illusion can be found in music. When you are introduced to a new song and listen to it a few times so that it becomes relevant to you, it is not uncommon to begin realizing that the song plays often on the radio or in public places. So, the next time you meet someone new and start to feel as though they are following you everywhere, fear not, you are just experiencing the frequency illusion. Unless, of course, this person has a crush on you and is practicing the mere exposure effect!



Conway, A., Cowan, N., Bunting, M. (2001). The cocktail party phenomenon revisited: The importance of working memory capacity. Psychonomic Bulletin & Review, 8 (2), 331-335. doi:https://link.springer.com/content/pdf/10.3758%2FBF03196169.pdf

Galbraith, R., Underwood, B. (1973). Perceived frequency of concrete and abstract words. Memory and Cognition, 1 (1), 56-60. https://doi.org/10.3758/BF03198068

Mendel, R., Traut-Mattausch, E., Jonas, E., Leucht, S., Kane, J. M., Maino, K., . . . Hamann, J. (2011). Confirmation bias: Why psychiatrists stick to wrong preliminary diagnoses. Psychological Medicine, 41(12), 2651-9. doi:http://dx.doi.org/10.1017/S0033291711000808

Zwicky, A.M. (2006). Why are we so illuded? Stanford University Psychology. doi:https://web.stanford.edu/~zwicky/LSA07illude.abst.pdf

  1. December 5th, 2019 at 14:47 | #1

    I really liked the way that you broke down the frequency illusion bias into two components: confirmation bias and selective attention. More importantly, your discussion on selective attention and how that impacts the frequency illusion made me think about what we discussed in class on memory and how memory is reconstructive and what we remember isn’t a scene by scene memory but bits of pieces that are then put together and filled in with prior knowledge and biases. In other words, when we fall victim of the frequency illusion bias, we think that we remember seeing someone or something more often than we actually have because we are focusing on attention on it and therefore when we reconstruct the memory, those things stand out the most. Or we might even think we saw someone when in fact we didn’t.

  2. December 7th, 2019 at 13:24 | #2

    I agree with Charlotte, I really like how you clearly defined and gave examples for the two principles behind the frequency illusion, confirmation bias and selective attention. Your definition of confirmation bias, that people actively seek ways to confirm their original beliefs, reminded me of weapon bias, which we discussed in class. In the weapon bias task, when people are shown a white face before being shown either a tool or a weapon, they are more likely to falsely recognize a weapon as a tool, and when shown a black face before a tool or weapon, are more likely to falsely recognize a tool as a weapon. They are ultimately trying to confirm their belief that black people are more like to have/use weapons than white people. Thinking about confirmation bias also made me realize that I very easily fall victim to it. I recently found out my friend voted for Trump, and looking back on it I’ve really been looking for signs that he isn’t as good of a guy as I thought…does he, as a white male, consider himself better than others? Is he ignorant toward the fact that I have a girlfriend? I’m sure he is the same good guy that I have always thought he was, but I have definitely been trying to fit him to my mold of a Trump supporter.

You must be logged in to post a comment.