The theory of signal detection was first introduced by Dutch scientist Adrianus (Barry) Michiel (Swets) in a groundbreaking paper in 1964. At that time, psychology was increasingly focusing on perception, thinking that the mind can directly perceive things.
This idea was strengthened by the fact that psychologists were starting to use experimental methods such as experiments and questionnaires to assess psychological phenomena.
However, Swets showed that this view is too simplistic. He pointed out that our perception is influenced by many factors, such as expectations, past experience, knowledge, and so on. All of these may play a role in what we “perceive” through our senses.
For example, if you expect your boss to come into the office late on Friday afternoon, you may “perceive” him or her coming into the office late due to your expectation. In reality, they may have arrived early and are working hard behind the desk while you are sleeping in your office chair.
This is what signal detection theory explains: how we perceive things based on our expectations and other factors.
History of signal detection theory
Signal detection theory (SDT) was first developed by Dutch scientist Adriaan de Groot in 1965. De Groot’s work was influenced by his study of chess, where he observed that players had to decide whether a given move would result in an advantage or disadvantage.
They had to discriminate the signals (advantageous moves) from the noise (disadvantageous moves). This led him to develop a theory of cognition that distinguished between these two components.
He extended this concept to all kinds of decision-making situations, including those involving perception. De Groot coined the terms “signal” and “noise” in this context, referring to what we perceive and what we do not perceive, respectively.
His SDT also hypothesized two additional factors that influence whether or not we detect a signal: our level of confidence that what we perceive is true and our motivation to act upon it.
Definition of probability
Probability is a measure of the likelihood that an event will occur. In marketing, this can be the likelihood that a customer will make a purchase or that a customer will respond to an advertisement.
Probability is calculated by the number of events divided by the total number of events and multiplicated by 100. For example, if there are 100 total events and 5 purchases, then the probability of a purchase is 5/100 = 0.5 = 50%.
Swets’s original focus was on detection in cases where detection was difficult. He defined detection as “the judgement that something has occurred”. In other words, it is determining whether or not something has occurred or whether something is true or false.
He explained that there are two main factors that influence our judgement of probability: prior experience and stimulus magnitude.
Three conditions of probability
In this article, we will focus on the second condition of probability: bias. We will discuss how bias affects our ability to detect signals and how we can account for bias in our decisions.
Bias is a tendency to think or act in a certain way that is consistent over time. A bias can be toward or against something, in regard to a certain process or judgment, or it can be a general attitude that influences everything else.
In psychology, the term “bias” is used to describe prejudiced thinking – such as racial prejudice. In this sense, bias is an extremely strong attitude that does not change easily.
In signal detection theory, the term “bias” refers to a systematic (i.e., occurring without exception) deviation from accuracy (i.e., detecting the signal) in one direction (i.e., underestimating or overestimating).
Signal to noise ratio (SNR)
Swets’ theory was not limited to performance in detection only. He also emphasized the importance of understanding how people perceive their performance in other areas, such as productivity.
How well you perceive your productivity is a factor in stress, anxiety, and overall wellbeing. A person with low self-esteem may feel like they are failing if they cannot get everything done on their list.
This is because they may not be paying close enough attention to the signals that their brain is giving them that they have done enough work.
They may be spending extended periods of time on tasks, but their internal sensors are telling them that they have completed enough work.
Since we cannot perceive our own internal sensors very well, it is important to pay attention to how much time you spend on tasks and things you complete.
Detecting a faint signal
Swets’s original focus was on how to detect a faint signal in noisy environments. This is also known as signal detection theory.
Signal detection theory deals with the process of deciding whether a signal is present or not. It addresses the questions: When do we think something is happening and when don’t we?
In this context, “signal” can refer to anything that may indicate that something important is happening. For example, a signal may be an indication that there is an oil spill, that someone has spotted a rare species of bird, or that there has been a rise in the stock market.
Regardless of what the signal may be, detection requires sensitivity to it, which in this context means the ability to recognize it. A “detection threshold” is set so that only signals above that level are recognized as such.
Example using swets’ original focus
Another example of where Swets’ theory has been applied is in the field of marketing. Marketers need to determine how to best allocate their marketing resources, or signals, in order to achieve the most sales with the resources they have.
They must determine how much money to spend on advertising, what types of advertising to use (radio ads, TV ads, etc.), and how many products to stock.
Again, just like in the example with the airport security staff, having too many ads will not help sales and may even hurt them. The same goes for having too much stock–if production cannot keep up with demand, then there is a loss of revenue.
The key is to find an equilibrium between detection (finding sales) and false alarm (spending money on non-buyers).
Applications of signal detection theory
Since its introduction, signal detection theory has been applied to a multitude of situations and scenarios. It has been used to explain a wide range of behaviors, from choosing relationships to choosing jobs.
It has been used to explain how people choose their relationships, whether that be romantic relationships or friendships. It has been used to explain how people choose their jobs, whether they stay in their current job or seek out another one.
It can even be applied to how people choose which foods they like and do not like. All of these decisions can be broken down into two main categories: liking or disliking something, and choosing something or nothing.
As mentioned earlier, Swets’ original focus was on improving the effectiveness of detecting signals in the work environment. This can apply to both employees and employers, as both need to understand what is beneficial for the other in order to effectively work together.
Summary
Swets’ (1964) signal detection theory was originally formulated to explain the performance of human observers in discriminating signals (target stimuli) from noise (non-target stimuli).
This theory has since been applied to many different situations and domains, including animal behavior, marketing, and finance. In finance, it has been used to explain performance in detecting economic signals such as inflation or rising prices.
Inflation is a signal that can affect the value of currency in the long run. Having an understanding of how people detect inflation can help predict how the value of the currency will change.
In animal behavior, it is used to understand how animals detect signals from other animals or their environment and respond appropriately. In marketing, it is used to determine how well products are received by consumers based on their perception of its quality.
Signal detection theory is a useful concept that can be applied in many situations.
Leave a Reply