Bias

This note is about bias that results from subjectivity, for example the many human biases, including the bias for finding narrative, meaning, and order where they do not exist.

Herzen wrote in an article entitled “Apropos of a Drama”:

"There is something about chance that is intolerably repellent to a free spirit…. He wants the misfortunes that overtake him to be predestined—that is, to exist in connection with a universal world order; he wants to accept disasters as persecutions and punishments: this allows him to console himself through submission or rebellion."

Herzen: the future is a variation improvised on the theme of the past.

Humans, as all living beings (cf. Biosemiotics) have inherent biases.

For example, apophenia, or patternicity, is a term/condition that refers to a tendency in humans where they see patterns where patterns do not exist.

Common Forms of Human Bias

  • confirmation bias: the tendency to search for, interpret, and remember information in a way that confirms one's preconceptions.
  • anchoring bias: the tendency to rely too heavily on the first piece of information encountered (the "anchor") when making decisions.
  • availability heuristic: overestimating the importance of information that is most readily available.
  • hindsight bias: the inclination to see events as having been predictable after they have already occurred.
  • self-serving bias: the tendency to attribute positive events to one's own disposition but attribute negative events to external factors.
  • ingroup bias: favouring members of one's own group over those in other groups.
  • status quo bias: the preference for the current state of affairs and resistance to change.
  • negativity bias: the tendency to give more weight to negative experiences or information than positive ones.
  • stereotyping: generalizing about a group of people in which identical characteristics are assigned to virtually all members of the group.
  • halo effect: the tendency to let an overall impression of a person influence specific judgments about them.

Bias in Technical Systems

  • algorithmic bias: when algorithms produce systematically prejudiced results due to erroneous assumptions in the machine learning process.
  • data bias: bias that occurs when the data used to train a model is not representative of the real-world scenario it is meant to reflect.
  • selection bias: bias introduced by the non-random selection of data, leading to a sample that is not representative of the population.
  • measurement bias: bias that arises from errors in data collection, leading to inaccurate or misleading data.
  • confirmation bias in AI: when AI systems are designed or trained in a way that confirms the developers' preconceptions.
  • interaction bias: bias that occurs when users interact with a system in a way that reinforces existing biases.
  • automation bias: the tendency to favour suggestions from automated decision-making systems and to ignore contradictory information made without automation.
  • survivorship bias: concentrating on the people or things that "survived" some process and overlooking those that did not due to their lack of visibility.
  • reporting bias: bias that occurs when certain outcomes or results are more likely to be reported than others.
  • group attribution bias: the tendency to generalize about a group based on the behaviour of a few individuals within that group.

References

Korteling, J. E. (Hans), Geke C. van de Boer-Visschedijk, Romy a. M. Blankendaal, Rudy C. Boonekamp, and Aletta R. Eikelboom. “Human- Versus Artificial Intelligence.” Frontiers in Artificial Intelligence 4 (2021): 622364. https://doi.org/10/gjrvcx.


Backlinks